You are on page 1of 341

Advances in Oil and Gas Exploration & Production

Hamid N. Alsadi

Seismic
Hydrocarbon
Exploration
2D and 3D Techniques
Advances in Oil and Gas Exploration &
Production

Series editor
Rudy Swennen, Department of Earth and Environmental Sciences,
K.U. Leuven, Heverlee, Belgium
The book series Advances in Oil and Gas Exploration & Production pub-
lishes scientic monographs on a broad range of topics concerning geo-
physical and geological research on conventional and unconventional oil and
gas systems, and approaching those topics from both an exploration and a
production standpoint. The series is intended to form a diverse library of
reference works by describing the current state of research on selected
themes, such as certain techniques used in the petroleum geoscience business
or regional aspects. All books in the series are written and edited by leading
experts actively engaged in the respective eld.
The Advances in Oil and Gas Exploration & Production series includes
both single and multi-authored books, as well as edited volumes. The Series
Editor, Dr. Rudy Swennen (KU Leuven, Belgium), is currently accepting
proposals and a proposal form can be obtained from our representative at
Springer, Dr. Alexis Vizcaino (Alexis.Vizcaino@springer.com).

More information about this series at http://www.springer.com/series/15228


Hamid N. Alsadi

Seismic Hydrocarbon
Exploration
2D and 3D Techniques

123
Hamid N. Alsadi
Data Processing Section
Ministry of Oil
Baghdad
Iraq

ISSN 2509-372X ISSN 2509-3738 (electronic)


Advances in Oil and Gas Exploration & Production
ISBN 978-3-319-40435-6 ISBN 978-3-319-40436-3 (eBook)
DOI 10.1007/978-3-319-40436-3

Library of Congress Control Number: 2016942906

Springer International Publishing Switzerland 2017


This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microlms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specic statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature


The registered company is Springer International Publishing AG Switzerland
Preface

The favorable reception of the rst edition of this book (seismic Exploration,
published by Birkhauser Verlag in 1980) stimulated my belief in the need of
an updated book that includes the advances in the techniques which have
taken place during the past three decades. In preparing the present updated
volume, I have taken into consideration the remarks and suggestions of the
users of the 1980 edition from both of the academic and industrial work
domains.
Since 1980, when the rst edition of this book was published, great
developments in the seismic exploration technology have taken place. These
developments have occurred in all of the three exploration phases: acquisi-
tion, processing, and interpretation techniques. The most prominent advances
which have taken place in these years are the widespread implementations
of the 3D surveying, pre-stack migration, and growing interpretation tech-
niques in both structural and stratigraphic exploration. As it is familiar with
the exploration geophysicists, this subject (seismic exploration) is fully dealt
with in many original and authentic internationally known text books. In this
publication, no new subjects added to those found in the other standard books
which are well known in the geophysical library. In fact, these and other
related scientic papers and research reports formed solid references for the
present book. There are, however, differences in the design and presentation
approach.
In its design, the book is intended to be a comprehensive treatise of the
seismic exploration tool, addressing audiences in both of the academic and
industrial establishments. It is made up of 12 chapters covering the basic
aspects of the seismic reflection exploration subject, starting with the basic
theory, followed by the applied data acquisition technology, and ending with
the processing and interpretation. In presenting the subject matter, emphasis
is made on the practical aspects of the subject, using clear and simplied
presentation, avoiding excessive descriptions and unnecessary lengthy
comments. Numerous illustration gures (>390 gures) have been used
throughout the book to aid in clarifying the concepts and procedures involved
in any standard seismic exploration survey. In this way, the book can be
considered as a very useful introductory teaching manual for university
students taking seismic reflection exploration as part of a postgraduate
course.

v
vi Preface

The chapters of the book are sequenced in the order of the activities
normally executed in a standard seismic exploration survey: eld acquisition,
processing, and interpretation. Chapter 1 is an introductory chapter in which
a brief historical note and short review of the geophysical exploration
methods are given. This is followed by four chapters covering the theoretical
aspect of the subject including basic principles and denitions of seismic
waves, with a special chapter assigned for the seismic wave propagation
velocity. The propagation phenomena, reflection, diffraction, transmission,
and refraction, are dealt with in Chaps. 4 and 5. The following two chapters
are devoted to the two main tools applied in seismic exploration, namely the
2D and 3D surveying techniques.
Due to its important role in understanding of the processing steps applied in
seismic data processing, a chapter (Chap. 8) is assigned solely for the seismic
signal. This chapter is structured on the theme of considering the seismic
reflection wavelet as a propagating signal in the same way as the electro-
magnetic signal is treated by the communication theory applied in electro-
magnetic wave propagation. Including a chapter on seismic signals, preceding
the processing chapters, is a feature by which this book has deviated from
other conventional publications. Data processing is presented under two
headings: processing tools (Chap. 9) and the normally applied processing
sequence (Chap. 10). Chapter 11 covers some specialized seismic exploration
tools sometimes used in support of the conventional seismic reflection and
refraction surveying. The book is concluded with Chap. 12 on interpretation.
I would like to express my gratitude to my wife Asira and my sons (Eng.
Majid, Ph.D., Eng. Muhannad, M.Sc., Thurayah, B.Sc., and Eng. Mahir,
B.Sc.), for their continuous support and help throughout the past three years.
My work in the writing of the book has incurred an additional burden to the
family especially during the abnormally difcult times, which our country
has experienced in the past twelve years.

Baghdad, Iraq Hamid N. Alsadi


December 2015
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Historical Review . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Elasticity and Seismic Waves (Bth 1979) . . . 1
1.1.2 Earthquake Seismology (Richter 1958). . . . . . 2
1.1.3 Exploration Seismology
(Telford et al. 1990) . . . . . . . . . . . . . . . . . . 2
1.1.4 Summary of Exploration Seismology History
(Sheriff and Geldart 1995) . . . . . . . . . . . . . . 3
1.2 The Geophysical Exploration Project . . . . . . . . . . . . . 3
1.3 Geophysical Exploration Methods . . . . . . . . . . . . . . . 4
1.3.1 The Seismic Method . . . . . . . . . . . . . . . . . . 4
1.3.2 The Gravity Method . . . . . . . . . . . . . . . . . . 4
1.3.3 The Magnetic Method . . . . . . . . . . . . . . . . . 6
1.3.4 The Electical Method. . . . . . . . . . . . . . . . . . 7
1.3.5 The Radioactivity Method . . . . . . . . . . . . . . 9
1.4 Oil Well Drilling. . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4.1 Drilling of the Exploration-Well . . . . . . . . . . 10
1.4.2 The Oil-Well Rotary Drilling . . . . . . . . . . . . 10
1.4.3 The Drill-Hole and Well Casing . . . . . . . . . . 11
1.5 Well Geophysical Logging . . . . . . . . . . . . . . . . . . . . 11
1.5.1 Electrical Logging . . . . . . . . . . . . . . . . . . . . 12
1.5.2 Radioactivity Logging . . . . . . . . . . . . . . . . . 14
1.5.3 Acoustic Logging . . . . . . . . . . . . . . . . . . . . 16
1.5.4 Log Interpretation . . . . . . . . . . . . . . . . . . . . 19
1.6 Latest Developments in Well Logging . . . . . . . . . . . . 19
1.6.1 Logging While Drilling Technique . . . . . . . . 20
1.6.2 Borehole Imaging Tools. . . . . . . . . . . . . . . . 20
1.7 Well Completion . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2 Seismic Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.1 The Fundamental Conditions . . . . . . . . . . . . . . . . . . 23
2.2 Theory of Elasticity . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2.1 Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.2 Strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.2.3 Common Types of Strain . . . . . . . . . . . . . . . 26
2.2.4 The Volume-Changing Strain . . . . . . . . . . . . 27
2.2.5 The Shape-Changing Strain . . . . . . . . . . . . . 28
2.2.6 The Cubical Dilatation . . . . . . . . . . . . . . . . . 28

vii
viii Contents

2.2.7 Stress-Strain Relationship . . . . . . . . . . . . . . . 29


2.2.8 Hookes Law for Isotropic Media . . . . . . . . . 29
2.2.9 The Elastic Moduli . . . . . . . . . . . . . . . . . . . 30
2.2.10 The Elastic Moduli Interrelationships . . . . . . . 32
2.3 Wave Motion Equation . . . . . . . . . . . . . . . . . . . . . . 32
2.3.1 One-Dimensional Scalar Wave Equation . . . . 33
2.3.2 The Scalar and Vector 3D Wave
Equations . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.3.3 Plane Waves. . . . . . . . . . . . . . . . . . . . . . . . 35
2.3.4 The P- and S-Waves . . . . . . . . . . . . . . . . . . 36
2.4 Classification of Common Elastic Waves . . . . . . . . . . 36
2.4.1 Body Waves. . . . . . . . . . . . . . . . . . . . . . . . 37
2.4.2 Surface Waves . . . . . . . . . . . . . . . . . . . . . . 38
2.4.3 Seismic Noise. . . . . . . . . . . . . . . . . . . . . . . 39
2.5 Propagation of Seismic Waves . . . . . . . . . . . . . . . . . 40
2.5.1 Elements of the Seismic Field. . . . . . . . . . . . 41
2.5.2 Concepts of Wave-Fronts and Rays . . . . . . . . 41
2.5.3 Huygens Principle . . . . . . . . . . . . . . . . . . . 42
2.5.4 The Concept of the Interface . . . . . . . . . . . . 42
2.5.5 Changes of Propagation Direction at
Interfaces . . . . . . . . . . . . . . . . . . . . . . .... 43
2.5.6 Wave Conversion at Interfaces . . . . . . . .... 44
2.5.7 Energy Partitioning and Zoeppritz
Equations . . . . . . . . . . . . . . . . . . . . . . .... 45
2.5.8 Amplitude Variation with Angle
of Incidence . . . . . . . . . . . . . . . . . . . . . . . . 46
2.6 Effect of the Medium on Wave Energy . . . . . . . . . . . 46
2.6.1 Geometrical Spreading . . . . . . . . . . . . . . . . . 47
2.6.2 Inelastic Attenuation . . . . . . . . . . . . . . . . . . 48
2.6.3 Seismic Wave Energy Measurement
and the DB Unit . . . . . . . . . . . . . . . . . .... 49
2.6.4 The Logarithmic Decrement . . . . . . . . . .... 49
2.6.5 Wave Dispersion. . . . . . . . . . . . . . . . . .... 49
3 The Seismic Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2 Factors Influencing Seismic Velocity . . . . . . . . . . . . . 53
3.2.1 Lithology Effect . . . . . . . . . . . . . . . . . . . . . 53
3.2.2 Elasticity and Density Effects . . . . . . . . . . . . 54
3.2.3 Porosity and Saturation Fluid Effects . . . . . . . 54
3.2.4 Depth and Geological Age Effects. . . . . . . . . 56
3.2.5 Overburden Pressure Effect . . . . . . . . . . . . . 56
3.2.6 Other Velocity-Affecting Factors. . . . . . . . . . 57
3.3 The Velocity Function . . . . . . . . . . . . . . . . . . . . . . . 58
3.4 Types of Velocity Functions . . . . . . . . . . . . . . . . . . . 59
3.4.1 Instantaneous Velocity . . . . . . . . . . . . . . . . . 59
3.4.2 Interval Velocity . . . . . . . . . . . . . . . . . . . . . 60
3.4.3 Average Velocity . . . . . . . . . . . . . . . . . . . . 60
3.4.4 Root Mean Square Velocity . . . . . . . . . . . . . 61
3.4.5 Stacking Velocity . . . . . . . . . . . . . . . . . . . . 62
Contents ix

3.4.6 The Apparent Velocity. . . . . . . . . . . . . . . . . 63


3.4.7 The Group and Phase Velocities . . . . . . . . . . 64
3.4.8 Representations of the Velocity Functions . . . 64
3.5 Velocity Determination Methods . . . . . . . . . . . . . . . . 64
3.5.1 The Well Velocity Survey . . . . . . . . . . . . . . 65
3.5.2 The Up-Hole Velocity Survey. . . . . . . . . . . . 66
3.5.3 Continuous Velocity Logging . . . . . . . . . . . . 66
3.5.4 (X2 T2 Method) . . . . . . . . . . . . . . . . . . . . 67
3.5.5 (T T) Method . . . . . . . . . . . . . . . . . . . . 67
3.5.6 Velocity Analysis Method . . . . . . . . . . . . . . 68
3.5.7 Seismic Velocity Inversion . . . . . . . . . . . . . . 69
3.6 Uses of the Seismic Velocity Data. . . . . . . . . . . . . . . 69
4 Seismic Wave Reflection and Diffraction . . . . . . . . . . . . .. 71
4.1 The Commonly-Recorded Seismic Events . . . . . . . .. 71
4.2 Wave Changes at Reflection Interface . . . . . . . . . . .. 71
4.2.1 Reflection Coefficient at Inclined
Incidence . . . . . . . . . . . . . . . . . . . . . . . . .. 72
4.2.2 Reflection Coefficient at Normal Incidence. .. 73
4.2.3 Geometry of Reflection from Dipping
Reflectors. . . . . . . . . . . . . . . . . . . . . . . . .. 74
4.2.4 Geometry of Reflection from Horizontal
Reflectors. . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.2.5 Reflection from Multiple Reflectors . . . . . . . . 75
4.3 The NMO and DMO Concepts . . . . . . . . . . . . . . . . . 76
4.3.1 The Normal Move-Out (NMO) Concept. . . . . 76
4.3.2 The Dip Move-Out (DMO) Concept . . . . . . . 79
4.4 The CDP, CRP, and CMP Concepts . . . . . . . . . . . . . 79
4.4.1 CDP, CRP, and CMP in Case
of Horizontal Reflector . . . . . . . . . . . . . . .. 80
4.4.2 CDP, CRP, and CMP in Case of Dipping
Reflector . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.5 The Seismic Wave Diffraction . . . . . . . . . . . . . . . . . 81
4.5.1 The Point-Diffractor . . . . . . . . . . . . . . . . . . 82
4.5.2 Diffraction from Terminating Reflectors . . . . . 83
4.5.3 Diffraction Seismic Image . . . . . . . . . . . . . . 83
4.5.4 Diffraction Travel-Time Function . . . . . . . . . 84
4.5.5 The Diffraction Hyperbola . . . . . . . . . . . . . . 85
4.5.6 Distortion Effects of Diffraction . . . . . . . . . . 86
4.5.7 The Exploding-Reflector Model . . . . . . . . . . 86
5 Seismic Wave Transmission and Refraction . . . . . . . . .... 89
5.1 Seismic Wave Transmission . . . . . . . . . . . . . . . .... 89
5.1.1 Transmission Coefficient at Normal
Incidence . . . . . . . . . . . . . . . . . . . . . . .... 89
5.1.2 The Two-Way Transmission Coefficient. .... 90
5.1.3 Attenuation Due to Reflection
and Transmission . . . . . . . . . . . . . . . . .... 90
5.1.4 Role of Transmission in Seismic
Exploration. . . . . . . . . . . . . . . . . . . . . .... 91
x Contents

5.2 Seismic Wave Refraction . . . . . . . . . . . . . . . . . . ... 92


5.2.1 Snells Law of Refraction. . . . . . . . . . . . ... 92
5.2.2 The Critical Refraction
and Head Wave Generation . . . . . . . . . . ... 95
5.2.3 Ray-Path Geometry and Travel-Time
Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.2.4 Refraction Travel-Time Function. . . . . . . . . . 96
5.2.5 The Delay-Time Concept . . . . . . . . . . . . . . . 98
5.2.6 Refraction in a Multi-layer Medium. . . . . . . . 99
5.2.7 Refraction in a Medium of Linear
Velocity Variation . . . . . . . . . . . . . . . . . ... 99
5.2.8 Refraction from a Faulted Refractor. . . . . ... 100
5.2.9 Applications of Seismic Wave Refraction . ... 100
6 2D Seismic Reflection Surveying . . . . . . . . . . . . . . . . . . . . 105
6.1 Reflection Surveying Concepts . . . . . . . . . . . . . . . . . 105
6.1.1 The Spread Configuration . . . . . . . . . . . . . . 105
6.1.2 The Seismic Trace. . . . . . . . . . . . . . . . . . . . 106
6.1.3 The Shotpoint . . . . . . . . . . . . . . . . . . . . . . . 107
6.1.4 The Seismic Profiling Technique. . . . . . . . . . 107
6.1.5 The Fold of Coverage . . . . . . . . . . . . . . . . . 108
6.1.6 The CDP Trace Gather . . . . . . . . . . . . . . . . 110
6.2 Seismic Reflection Data Acquisition . . . . . . . . . . . . . 113
6.2.1 Seismic Energy Sources . . . . . . . . . . . . . . . . 113
6.2.2 Seismic Detectors . . . . . . . . . . . . . . . . . . . . 119
6.2.3 The Seismic Data Recording. . . . . . . . . . . . . 121
6.2.4 Data Playback and Display. . . . . . . . . . . . . . 126
6.3 Seismic Noise Characterization and Attenuation . . . . . 126
6.3.1 Random Noise Attenuation. . . . . . . . . . . . . . 127
6.3.2 Coherent Noise Attenuation . . . . . . . . . . . . . 127
6.4 Field Measures for Signal Enhancement . . . . . . . . . . . 131
6.4.1 The Noise Test . . . . . . . . . . . . . . . . . . . . . . 131
6.4.2 The Experimental Shooting . . . . . . . . . . . . . 134
6.4.3 Determination of the LVL Properties . . . . . . . 134
6.5 The Seismic Field Crew . . . . . . . . . . . . . . . . . . . . . . 136
7 3D Seismic Reflection Surveying . . . . . . . . . . . . . . . . . . . . 139
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
7.1.1 Nature Is Three Dimensional . . . . . . . . . . . . 139
7.1.2 1D-2D-3D Terminology . . . . . . . . . . . . . . . . 140
7.1.3 Limitations of 2D Seismic Surveying. . . . . . . 141
7.1.4 Merits of the 3D Technique . . . . . . . . . . . . . 143
7.1.5 Survey Preplanning . . . . . . . . . . . . . . . . . . . 146
7.1.6 Cost Considerations of 3D Surveys . . . . . . . . 146
7.2 Definitions and Basic Principles . . . . . . . . . . . . . . . . 148
7.2.1 Definition of 3D Surveying . . . . . . . . . . . . . 148
7.2.2 The 3D Spread-Geometry. . . . . . . . . . . . . . . 149
7.2.3 Concept of the Trace Azimuth . . . . . . . . . . . 149
Contents xi

7.2.4 The CDP Bin and Bin Attributes. . . . . . . . . . 150


7.2.5 The Surface and Sub-surface Coverage . . . . . 150
7.2.6 The Seismic Data Volume . . . . . . . . . . . . . . 151
7.3 3D Field Data Acquisition . . . . . . . . . . . . . . . . . . . . 153
7.3.1 Types of Marine 3D Spreads . . . . . . . . . . . . 153
7.3.2 Types of Land 3D Spreads . . . . . . . . . . . . . . 155
7.3.3 The Swath Shooting Technique. . . . . . . . . . . 156
7.3.4 Types of Templates Used in Swath
Shooting . . . . . . . . . . . . . . . . . . . . . . . . . . 156
7.3.5 The Loop Shooting Technique . . . . . . . . . . . 159
7.3.6 The 3D Survey Design . . . . . . . . . . . . . . . . 159
7.4 Processing of 3D Data . . . . . . . . . . . . . . . . . . . . . . . 164
7.4.1 The Field Data Processing . . . . . . . . . . . . . . 166
7.4.2 The Final Data Processing . . . . . . . . . . . . . . 166
7.4.3 3D Data Displays . . . . . . . . . . . . . . . . . . . . 166
8 The Seismic Reflection Signal . . . . . . . . . . . . . . . . . . . . . . 169
8.1 Definition of the Seismic Signal . . . . . . . . . . . . . . . . 169
8.1.1 The Seismic Signal Parameters . . . . . . . . . . . 170
8.1.2 The Seismic Signal Spectrum . . . . . . . . . . . . 170
8.1.3 Comparison with the Radar Signal . . . . . . . . 172
8.1.4 The Seismic Reflection-Signal Changes . . . . . 172
8.2 The Seismic Trace. . . . . . . . . . . . . . . . . . . . . . . . . . 173
8.2.1 The Convolutional Model of the Seismic
Trace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
8.2.2 The Synthetic Seismogram . . . . . . . . . . . . . . 175
8.2.3 The Digital Seismic Trace . . . . . . . . . . . . . . 176
8.3 The Wavelet Concept . . . . . . . . . . . . . . . . . . . . . . . 177
8.3.1 Definition of the Seismic Wavelet . . . . . . . . . 177
8.3.2 Energy and Delay Properties of Wavelets. . . . 178
8.4 Sampling and the Digital Function . . . . . . . . . . . . . . 178
8.4.1 The Digital Function . . . . . . . . . . . . . . . . . . 179
8.4.2 The Sampling Process . . . . . . . . . . . . . . . . . 179
8.4.3 Representation Methods of the Digital
Function . . . . . . . . . . . . . . . . . . . . . . . . . . 180
8.5 The Sampling Theorem and Aliasing . . . . . . . . . . . . . 181
8.5.1 The Sampling Theorem . . . . . . . . . . . . . . . . 181
8.5.2 The Aliasing Phenomenon . . . . . . . . . . . . . . 182
8.5.3 The Aliasing Frequency Computation . . . . . . 183
8.5.4 Effect of Sampling on Signal Spectrum . . . . . 185
8.5.5 Aliasing in the Frequency Domain . . . . . . . . 185
8.5.6 The Remedy for the Aliasing Effect . . . . . . . 186
8.6 Signal Resolution and Resolution Power . . . . . . . . . . 187
8.6.1 Vertical Resolution of Seismic Signals. . . . . . 187
8.6.2 Horizontal Resolution of Seismic Signals . . . . 189
8.6.3 Fresnel Zone Formula . . . . . . . . . . . . . . . . . 189
xii Contents

8.7 The Common Numbering Systems . . . . . . . . ...... 190


8.7.1 Numbering System Concept . . . . . . . ...... 191
8.7.2 Applications of the Concept . . . . . . . ...... 191
8.7.3 Counting in the Different Numbering
Systems . . . . . . . . . . . . . . . . . . . . . ...... 195
9 The Seismic Processing Tools . . . . . . . . . . . . . . . . . . . . . . 197
9.1 The Seismic Processing Tools. . . . . . . . . . . . . . . . . . 197
9.1.1 The Sine Function . . . . . . . . . . . . . . . . . . . . 197
9.1.2 The Degrees-Radians Relationship. . . . . . . . . 198
9.1.3 Parameters of the Sine Function . . . . . . . . . . 198
9.1.4 The Frequency Concept . . . . . . . . . . . . . . . . 199
9.1.5 The Phase Concept . . . . . . . . . . . . . . . . . . . 200
9.1.6 Temporal and Spatial Frequencies . . . . . . . . . 200
9.1.7 Propagating Sine Wave . . . . . . . . . . . . . . . . 202
9.2 Fourier Analysis and Concept of Spectra . . . . . . . . . . 203
9.2.1 Fourier Series . . . . . . . . . . . . . . . . . . . . . . . 203
9.2.2 Gibbs Phenomenon . . . . . . . . . . . . . . . . . . 205
9.2.3 Fourier Transform . . . . . . . . . . . . . . . . . . . . 205
9.3 Concept of the Frequency Spectrum . . . . . . . . . . . . . 206
9.3.1 The Line Spectrum . . . . . . . . . . . . . . . . . . . 207
9.3.2 The Continuous Spectrum . . . . . . . . . . . . . . 207
9.3.3 The Fourier Power Spectrum . . . . . . . . . . . . 207
9.3.4 The Two-Domain Concept . . . . . . . . . . . . . . 208
9.3.5 Spectrum of the Rectangular Pulse . . . . . . . . 209
9.3.6 Spectrum of the Spike Pulse . . . . . . . . . . . . . 210
9.3.7 The Dirac-Delta Function . . . . . . . . . . . . . . . 210
9.3.8 Frequency Limits of Fourier Spectra . . . . . . . 211
9.4 The Phase Spectrum . . . . . . . . . . . . . . . . . . . . . . . . 212
9.4.1 The Zero-Phase Spectrum. . . . . . . . . . . . . . . 212
9.4.2 The Linear-Phase Spectrum . . . . . . . . . . . . . 213
9.4.3 The Constant-Phase Spectrum . . . . . . . . . . . . 213
9.5 Spectra of Observational Data. . . . . . . . . . . . . . . . . . 214
9.5.1 Truncation Effect and Windowing . . . . . . . . . 215
9.5.2 The Rectangular Window (Box-Car) . . . . . . . 216
9.5.3 Triangular Window (Bartlett) . . . . . . . . . . . . 217
9.6 Correlation Functions. . . . . . . . . . . . . . . . . . . . . . . . 218
9.6.1 Cross Correlation Function, C12(). . . . . . . . . 218
9.6.2 Autocorrelation Function C11() . . . . . . . . . . 218
9.6.3 Properties and Applications of Correlation
Functions . . . . . . . . . . . . . . . . . . . . . . . ... 219
9.6.4 Correlation Functions in the Frequency
Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
9.7 Convolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
9.8 Deconvolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
9.8.1 Deconvolution Main Objectives . . . . . . . . . . 222
9.8.2 Spiking (Whitening) Deconvolution. . . . . . . . 223
9.8.3 Gapped Deconvolution. . . . . . . . . . . . . . . . . 224
9.8.4 Noise Behaviour Under Deconvolution . . . . . 225
Contents xiii

9.9 Frequency Filtering . . . . . . . . . . . . . . . . . . . . ..... 226


9.9.1 Definition of the Filtering Process . . . . ..... 226
9.9.2 Role and Objectives of Filtering
in Seismic Exploration . . . . . . . . . . . . . . . . . 227
9.9.3 Concept of the Linear System . . . . . . . . . . . . 228
9.9.4 The Filter Impulse Response. . . . . . . . . . . . . 229
9.9.5 Mechanism of the Filtering Process . . . . . . . . 230
9.9.6 Filtering in the Two Domains . . . . . . . . . . . . 230
9.9.7 Types of Filters. . . . . . . . . . . . . . . . . . . . . . 231
9.9.8 Application of the Band-Pass Filter . . . . . . . . 231
9.9.9 Filter Shape (Side-Lobes Problem) . . . . . . . . 232
9.9.10 Effect of the Filter Length . . . . . . . . . . . . . . 233
9.9.11 Effect of the Filter Spectrum Shape . . . . . . . . 233
9.9.12 Minimization of the Truncation Effect . . . . . . 235
9.9.13 Filter Normalization. . . . . . . . . . . . . . . . . . . 235
9.9.14 Time-Variant Filtering . . . . . . . . . . . . . . . . . 236
9.9.15 The Filter Tests. . . . . . . . . . . . . . . . . . . . . . 237
9.9.16 Phase Removing Filtering . . . . . . . . . . . . . . 238
9.10 Frequency-Wave Number (FK) Filtering . . . . . . . . . . 238
9.10.1 The t-x and F-K Domains . . . . . . . . . . . . . . 238
9.10.2 Application of the F-K Filtering . . . . . . . . . . 239
9.11 Seismic Trace Equalization. . . . . . . . . . . . . . . . . . . . 240
9.11.1 The Need for Trace Equalization. . . . . . . . . . 240
9.11.2 Time-Invariant Equalization . . . . . . . . . . . . . 242
9.11.3 Time-Variant Equalization . . . . . . . . . . . . . . 242
9.12 Trace-Samples Manipulations . . . . . . . . . . . . . . . . . . 243
10 Processing of Seismic Reflection Data . . . . . . . . . . . . . . . . 245
10.1 General Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 245
10.1.1 Main Processing Objectives . . . . . . . . . . . . . 246
10.1.2 Role of Processing in Seismic Exploration . . . 247
10.1.3 Main Processing Activities . . . . . . . . . . . . . . 247
10.1.4 The Processing Sequence . . . . . . . . . . . . . . . 248
10.2 Data Re-organization . . . . . . . . . . . . . . . . . . . . . . . . 250
10.2.1 Seismic Data Copying, Reformatting,
and Loading . . . . . . . . . . . . . . . . . . . . . . . . 250
10.2.2 Data De-multiplexing. . . . . . . . . . . . . . . . . . 250
10.2.3 Sweep Removal . . . . . . . . . . . . . . . . . . . . . 251
10.2.4 Vertical Stacking. . . . . . . . . . . . . . . . . . . . . 252
10.2.5 Geometry Data and Field Statics Loading. . . . 252
10.2.6 Trace Header Assignment. . . . . . . . . . . . . . . 253
10.2.7 Data Re-sampling . . . . . . . . . . . . . . . . . . . . 253
10.3 Pre-stack Processing . . . . . . . . . . . . . . . . . . . . . . . . 253
10.3.1 Trace Editing . . . . . . . . . . . . . . . . . . . . . . . 253
10.3.2 Noise Attenuation . . . . . . . . . . . . . . . . . . . . 254
10.3.3 True Amplitude Recovery . . . . . . . . . . . . . . 254
10.3.4 Field Static Correction . . . . . . . . . . . . . . . . . 256
10.3.5 Application of Deconvolution . . . . . . . . . . . . 259
10.3.6 Trace Equalization. . . . . . . . . . . . . . . . . . . . 262
10.3.7 CMP-Sorting . . . . . . . . . . . . . . . . . . . . . . . 263
xiv Contents

10.3.8 NMO Correction . . . . . . . . . . . . . . . . . . . . . 264


10.3.9 Trace Muting . . . . . . . . . . . . . . . . . . . . . . . 267
10.4 Stack and Post-stack Processing . . . . . . . . . . . . . . . . 268
10.4.1 CMP Stacking of Seismic Traces . . . . . . . . . 269
10.4.2 CMP Stacking . . . . . . . . . . . . . . . . . . . . . . 269
10.4.3 Vertical Stacking. . . . . . . . . . . . . . . . . . . . . 270
10.4.4 Horizontal Stacking . . . . . . . . . . . . . . . . . . . 270
10.4.5 Diversity Stacking . . . . . . . . . . . . . . . . . . . . 271
10.4.6 The Stack Section . . . . . . . . . . . . . . . . . . . . 271
10.5 Seismic Migration . . . . . . . . . . . . . . . . . . . . . . . . . . 272
10.5.1 Distortions in the Stack Section . . . . . . . . . . 272
10.5.2 Problem Statement . . . . . . . . . . . . . . . . . . . 273
10.5.3 Geometry of the Distortions . . . . . . . . . . . . . 275
10.5.4 Relationship of Post-migration Dip
to Pre-migration Dip . . . . . . . . . . . . . . . . . . 275
10.5.5 Types of Migration Techniques. . . . . . . . . . . 276
10.5.6 Time Versus Depth Migration. . . . . . . . . . . . 281
10.5.7 Migration Parameters . . . . . . . . . . . . . . . . . . 282
10.5.8 Migration Techniques in Future . . . . . . . . . . 284
10.6 Application of Filtering and Equalization . . . . . . . . . . 285
10.6.1 Frequency Filtering . . . . . . . . . . . . . . . . . . . 285
10.6.2 Equalization . . . . . . . . . . . . . . . . . . . . . . . . 285
10.6.3 Data Storage and Display . . . . . . . . . . . . . . . 286
10.7 Parameter Optimization . . . . . . . . . . . . . . . . . . . . . . 286
10.7.1 Velocity Analysis . . . . . . . . . . . . . . . . . . . . 286
10.7.2 Residual-Static Analysis. . . . . . . . . . . . . . . . 287
10.7.3 Deconvolution Analysis . . . . . . . . . . . . . . . . 289
10.7.4 Filter Analysis . . . . . . . . . . . . . . . . . . . . . . 289
10.7.5 Equalization Analysis . . . . . . . . . . . . . . . . . 290
10.7.6 Mute Analysis . . . . . . . . . . . . . . . . . . . . . . 290
11 Extra Exploration Tools . . . . . . . . . . . . . . . . . . . ....... 291
11.1 Seismic Exploration with S-waves . . . . . . . . ....... 291
11.1.1 Shear-Waves Generation . . . . . . . . ....... 291
11.1.2 Role of the Shear-Wave Velocity
in Interpretation . . . . . . . . . . . . . . . . . . . . . 292
11.2 Vertical Seismic Profiling (VSP) . . . . . . . . . . . . . . . . 292
11.2.1 Field Set-up and Types of VSP. . . . . . . . . . . 292
11.2.2 The VSP Field Operation . . . . . . . . . . . . . . . 293
11.2.3 The VSP Seismic Record . . . . . . . . . . . . . . . 293
11.2.4 The VSP Data Processing . . . . . . . . . . . . . . 294
11.3 Seismic Tomography . . . . . . . . . . . . . . . . . . . . . . . . 295
11.3.1 The Forward and Inverse Modeling
Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . 295
11.3.2 Types of Seismic Tomography . . . . . . . . . . . 296
11.3.3 Interpretation of Tomography Data . . . . . . . . 296
11.4 4D (Time-Lapse) Surveying . . . . . . . . . . . . . . . . . . . 297
11.5 Multi-component (3C and 4C) Recording . . . . . . . . . . 297
11.6 Passive Seismic Surveying . . . . . . . . . . . . . . . . . . . . 298
11.7 Tau-P Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Contents xv

11.8 Amplitude Variation with Offset (AVO). . . . . . . . . . . 299


11.9 Seismic Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . 299
12 Interpretation of Seismic Reflection Data . . . . . . . . . . . . . . 301
12.1 Scope and Objectives. . . . . . . . . . . . . . . . . . . . . . . . 301
12.2 The Seismic Interpretation Tools . . . . . . . . . . . . . . . . 301
12.3 The Seismic Structural Interpretation . . . . . . . . . . . . . 302
12.3.1 Structural Interpretation Sequence . . . . . . . . . 302
12.3.2 The Seismic Structural Features . . . . . . . . . . 304
12.4 The Seismic Stratigraphic Interpretation . . . . . . . . . . . 306
12.4.1 Role of the Seismic Wavelet
in Stratigraphic Interpretation . . . . . . . . . . . . 306
12.4.2 Basic Stratigraphic Concepts. . . . . . . . . . . . . 307
12.4.3 Reflection Configuration Patterns . . . . . . . . . 308
12.4.4 Seismo-Stratigraphic Features . . . . . . . . . . . . 310
12.4.5 Stratigraphic Interpretation Sequence . . . . . . . 315
12.5 Direct Hydrocarbon Detection and Seismic
Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 316
12.5.1 Hydrocarbon Indicators . . . . . . . . . . ...... 316
12.5.2 Seismic Attributes . . . . . . . . . . . . . . ...... 317

References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Introduction
1

1.1 Historical Review 1.1.1 Elasticity and Seismic Waves


(Bth 1979)
The science of Seismic Exploration is in essence
based on the elastic-wave phenomenon and its The rst scientist who investigated the elastic
applications in exploring subsurface geological behavior of bodies is Galileo (15641642), who,
structures with what they contain from minerals, in 1638 investigated the elastic behavior of a
water, and hydrocarbon deposits. This phe- loaded beam attached at one end to a wall. In
nomenon depends largely on the elastic properties 1660, Hooks law was established. This law
of the medium within which the wave eld pre- forms the basis for the mathematical theory of
vails. The earliest discoveries were, in fact, made Elasticity.
in the eld of elasticity property and its role in In 1828, the French scientist, S. D. Poisson
generation and propagation of the elastic waves. found that two types of waves (the longitudinal
Following to these activities, intensive studies and transverse elastic waves) can travel through
were conducted on the elastic waves generated by an elastic medium with propagation velocities,
earthquakes which are considered to be natural VP and VS where the ratio (VP/VS) is equal to
energy sources causing seismic waves. In the last square root of 3. This result was later conrmed
phase of the research development, the seismic by Stokes in England in 1849, who labeled the
phenomenon created by natural earthquakes was two wave-types by P for the longitudinal and S
simulated using articial seismic energy sources for the transverse waves. Stokes also showed that
directed towards geological exploration. For eco- the applied stress onto a body can be analyzed
nomic benets, the seismic exploration technol- into normal component causing compression (or
ogy was further developed as a specialized method tension) and tangential component causing shear
used in mineral and petroleum exploration. deformation. The two modulii which Stokes
In view of this piece of information, it is dened to express these effects (body resistances
possible to review the development of the seis- to stresses) are what is now known as the modulii
mic exploration under the three historical phases: of compressibility and rigidity of materials.
Historical Development of Elasticity and In 1887, Lord Rayleigh, in England, discov-
Seismic Waves ered a type of elastic waves which can propagate
Historical Development of Earthquake over the surface of a medium with velocity (VR)
Seismology which is lower than VS in the same medium. This
Historical Development of Exploration was followed by another similar discovery, in
Seismology 1911, of another type of surface waves: the Love

Springer International Publishing Switzerland 2017 1


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_1
2 1 Introduction

waves, named after its discoverer, the English- In recent times, we may add Agadir-Morocco
man A. E. H. Love. in 1960, Kobe-Japan in 1995, Fukushima-Japan
in 2011, and Katimando-Nipal in 2015.

1.1.2 Earthquake Seismology 1.1.3 Exploration Seismology


(Richter 1958) (Telford et al. 1990)
Very large number of earthquakes is continu-
The normal mineral and oil exploration seis-
ously occurring on various parts of the Earth. If
mology is, in principle, the same as earthquake
small shocks are counted, 100 000 shocks a year
seismology. The main difference is in the energy
is considered to be a conservative estimate
source used to generate the seismic waves.
(Richter 1958, p. 8). These events have furnished
The natural energy source used in earthquake
most of the fundamental information concerning
seismology is replaced by the articial
seismic waves and other aspects of the earth-
mechanically-generated seismic energy. Broadly
quake seismology. Pre-historic earthquakes were
speaking, earthquake seismology was developed
reported in terms of their destruction effects and
during the rst half of the nineteenth century,
loss of lives. In Gutenbergs handbook of Geo-
whereas the oil-prospecting seismic-techniques
physics (Handbuch der Geophysik, volumes 14,
were developed during the rst half of the
Berlin 1929), valuable information on earthquake
twentieth century, using the refraction and
occurrence in pre historic times were indorsed. In
reflection techniques.
particular, he mentioned in volume-4 (pp. 803
The tragic incident of the Titanic ship that was
804), that pre-historic destructive earthquakes
hit and sunk by an iceberg in 1912, has moti-
were reported to have occurred at different places
vated R. Fessenden (Canadian) to use the phe-
of Mesopotamia. In this region four destructive
nomenon of reflection of acoustic waves in
earthquakes have occurred within the period
detecting icebergs. He also worked on using the
(1260550) BC.
principle in detecting submarines during the First
In the year (132 AD), the Chinese philosopher
World War (19141918). At about the same
Chang Heng devised a detection instrument to
time, L. Minntrop (German) devised a seismo-
detect the arriving earthquake impulse marking the
graph with which he succeeded to detect salt
rst step in instrumental seismology (Bullen
domes in Germany. With his newly founded
1965). However, the proper documentation and
company (Seismos), the rst refraction shooting
scientic investigations of earthquakes started
was conducted. This refraction-based survey
with the introduction of the detection instrument
resulted in 1924 in the discovery of the Orchard
(the seismograph) at the end of the 19th century. In
Dome southwest of Houston, Texas, USA.
1880, Grey, Milne, and Ewing (working in Japan)
Refraction exploration was then extensively
developed the rst scientic seismograph-based
applied until 1930 when the method gave way to
analyses of earthquake-generated seismic waves.
the more effective method of reflection shooting.
The main large-magnitude historic earthquakes
Based on Fessenden work in 1913 and Karcher
which are scientically described (based on seis-
application efforts in Oklahoma, USA, in the early
mograph data) are the following:
1920s, reflection seismology was developed into
Lisbon-Portugal in 1755, an efcient exploration tool. After 1930, reflection
Cutch- India in 1819, exploration was used on wide commercial scale.
Naples-Italy in 1819, A new era in the method development was the
Mino Owari-Japan in 1891, introduction in the early 1950s of the analogue
San Francisco-USA in 1906, data recording and playback data processing. In
Kwanto-Japan in 1923. 1953, the analogue magnetic tape recording and
1.1 Historical Review 3

processing was introduced. About 10 years later Year Nature of development


the digital systems were introduced and began to
1954 Introduction of the continuous velocity
replace the analogue systems rather rapidly. One logging
of the important features of the magnetic tape 1963 Introduction of the digital seismic
recording was the ability of data storage, playback, data-recording
and processing for the purpose of the signal 1965 Introduction of the Air-gun as seismic energy
enhancement of the reflection seismic event. source
1972 Bright spot and direct hydrocarbon detection
techniques
1.1.4 Summary of Exploration 1975 Introduction of Seismic Stratigraphy
Seismology History 1976 3D seismic reflection surveying, became in
(Sheriff and Geldart common use
1995) 1985 Introduction of the interpretation workstation
1990 Application of the sattelite global positioning
The main historical developments, the seismic system (GPS)
exploration method has witnessed can be sum-
marized as follows: A full coverage and detailed account on the
historical development of the exploration seis-
Year Nature of development mology is given by Sheriff and Geldart (1995,
1913 Invention of instruments (by R. Fessenden) to pp. 332).
record acoustic waves in water to detect
icebegs. His patent (issued in 1917) was the
rst in USA, on application of seismic waves
for exploration
1.2 The Geophysical Exploration
1914 Invention of portable seismographs (by Project
L. Minntrop) to record seismic waves
generated by articial explosions. These
detectors ware used to locate enemy artillery A geophysical exploration-project normally fol-
during World War-1. He was awarded a lows a sequence of activities starting with eld
US patent for this technique in 1923 work and ending up with production of the
1923 Use of rst seismic refraction shooting. subsurface geological model of the project area.
Conducted in the Gulf Coast area, resulted in An exploration project passes through a sequence
1924 in the discovery of the Orchard Dome,
Texas, USA of three phases, normally executed in sequence.
The starting phase of a survey is collecting the
1926 Use of seismic reflection shooting in the state
of Oklahoma, USA eld raw data through standard eld procedures
1927 Use of rst well velocity survey
by which the geophysical measured values are
recorded, usually on magnetic tapes. After com-
1933 Use of multiple geophone per seismic channel
(the geophone group) pleting the acquisition work, the recorded data
are passed on to a processing centre where it is
1936 First reproducible photographic recording
developped by F. Rieber subjected to certain processing steps for purpose
1944 Introduction of large-scale marine surveying of certain corrections and enhancing of the geo-
physical signal. The third and last phase of the
1952 Introduction of the analogue magnetic
recording and play-back recording project is to interpret the nal processed data to
1953 Introduction of the Vibroseis and
extract the subsurface geological model of the
Weight-dropping as energy generators area under exploration. The three phases of
(continued) exploration work is summarized in Fig. 1.1.
4 1 Introduction

Field Data Acquisition Data Processing Data Interpretation

Fig. 1.1 The three phases of the seismic exploration project

Detailing of this technique is the subject of this


1.3 Geophysical Exploration monograph.
Methods There are two main techniques involved by the
seismic method; the reflection and the refraction
The main exploration methods employing geo- methods (Fig. 1.2).
physical principles are: In a seismic experiment, there are generally
(i) Seismic four types of wave arrivals that can be detected
(ii) Gravity by a receiver placed at an offset-distance from the
(iii) Magnetic seismic source. These are:
(iv) Electrical Reflection Arrivals
(v) Radioactivity Refraction Arrivals
(vi) Electromagnetic Diffraction Arrivals
Seismic method is by far the most widely Direct (Transmission) Arrivals
applied method in the exploration of oil and gas These four types of seismic wave arrivals can be
deposits. Gravity, magnetic, electrical, and ra- considered to be the main seismic exploration-tools
dioactivity methods provide support data in the involved in any seismic exploration project.
oil exploration and also applied as the main tools
for other mineral exploration. Geophysical tech-
niques which are using the Earth geophysical 1.3.2 The Gravity Method
potential elds (as gravity and magnetic elds)
are often called (Potential Methods). In this method the acceleration due to the Earth
These methods are briefly dened as follows: gravitational force is measured at a grid of points
distributed over the survey area. The observed
1.3.1 The Seismic Method gravity values are reduced to what they would
be, had they been measured at a xed datum
This method is based on generating seismic waves level (normally xed at the sea level). The
by a mechanical energy source at a point on, or just so-corrected gravity values are contoured to
below, the ground surface and recording the arri- show the variation of the gravity values (in
val at another surface point of the reflected (or acceleration units) throughout the area.
refracted) waves. From the travel-time measure- It is known that gravity changes depend on
ments of these waves and wave motion-velocity, changes in density as well as on the depth of the
the structural variation of subsurface geological causing geological anomaly. Thus, as the density
layers are mapped. Under favorable conditions, increases (relative to the host medium), the
study of the seismic wavelet can provide infor- gravity attraction (or acceleration value) of a rock
mation on the stratigraphic nature and hydrocar- mass at a given depth increases. If, however, a
bon contents of the traversed rocks. The raw constant-density rock mass exists at different
seismic data (normally recorded on magnetic depths, the corresponding gravity values are
tapes), are passed through a sequence of process- different. Gravity values of deeper rocks are less
ing steps followed by a set of certain interpretation than those for shallower rocks. A typical example
procedures in order to obtain the nal result, which for this case is the gravity changes normally
is a model of the subsurface geological structure. exhibited over an anticline, where the gravity
1.3 Geophysical Exploration Methods 5

(a) source receivers

seismic
reflection depth
raypath
reflector

(b) source receivers

seismic
refraction depth
raypath
reflector

Fig. 1.2 Essence of the two seismic techniques; reflection (a) and refraction (b)

change (called the gravity anomaly) is greater The measuring instruments, normally used in
over the anticline axis falling to lower values gravity surveying, are called the gravity meters
over the flanks. A structural feature showing (or gravimeters). These are designed to measure
relative density-deciency (such as a salt dome), gravity variations rather than absolute gravity
gives an inverted bell-shaped anomaly normally values. These instruments are capable to measure
referred to as (negative anomaly) which is of gravity changes to less than a tenth of a milligal.
opposite shape-changes to that exhibited by the The gravity unit (the gal) is dened to be equal to
relatively denser-rock geological anomaly (posi- 1 cm/s2. Thus the gravimeter is capable of mea-
tive anomaly). The principle is shown in Fig. 1.3. suring gravity variations to within about

positive gravity anomaly negative gravity anomaly

0 0

Sea level

geological anomaly
1 2 1 2

Fig. 1.3 Dependence of the algebraic sign of surplus where body density (q2) is greater than that of the
gravity-anomaly on whether the contrast is surplus or host medium (density, q1), and vice versa, given (q2 > q1)
deffeciency. Positive sign is given to case of density
6 1 Introduction

ten-millionth of the Earths total gravity eld


which is about 9.8 k gal.
Gravity surveying is ideally suited to map
subsurface rock layers or mineral deposits that S(-) N (+)
show depth-variations such as folded strata, or
density-changes as in cases of salt domes, heavy
mineral deposits, and subsurface cavities.
Because of the two factors affecting the measured
Fig. 1.4 Magnetic poles of a magnetic body, are two
gravity change (density and depth), interpretation points near magnet ends to which force lines are pointing;
of the gravity data suffers from ambiguity in the the North (+N) and South (S) poles. Arrows indicate the
determination of the real geological anomaly conventional movement-direction of a unit positive pole
placed in the magnetic eld
responsible of creating the gravity anomaly. For
this reason an additional tool or information is
needed to determine the true causal geological reason, magnetic surveying is normally carried
anomaly. out to explore magnetized materials, such as
Details of this method can be found in many iron-ore deposits, igneous intrusions, and sur-
standard geophysical publications. A simplied faces of basement rocks. Rocks which have
monograph on the subject that can be referred to high-susceptibility mineral contents, acquire
is Alsadi and Baban 2014. magnetic intensity by magnetic induction pro-
cess. This leads to creating of a new magnetic
eld which is added to the already existing Earth
1.3.3 The Magnetic Method ambient magnetic eld. The eld magnetic sur-
vey aims at recording of the measured magnetic
As often stated in the geophysical literature, the eld which is then subjected to processing and
magnetic method is considered to be the oldest isolating the magnetic anomalies making them
method applied in geophysical exploration. The ready for interpretation. The value of the mag-
Magnetic Phenomenon is force of attraction or netic anomaly is proportional to the intensity of
repulsion due to electron arrangement in certain magnetization caused by the geological anomaly.
substances having magnetic properties. The After certain processing measures these anoma-
magnetic body (magnet) is considered to be lies are interpreted to reveal the causal geological
made up of small parts called (magnetic structure.
domains) which are lined up in the same direc- The unit of magnetic-eld measurement is the
tion. Unlike the case of gravity, the magnetic Oersted, where one Oersted is equal to one dyne
body has always two poles (dipole-body); one per unit magnetic pole. In practice, another
has attraction effect and the other has repulsion smaller unit (gamma = 105 Oersted) is used.
effect. An end-point of a long bar magnet can be More recently an SI unit called nanotesla where
considered as an isolated magnetic pole for cer- 1 nanotesla is equal to 1 gamma. The SI (Systeme
tain computational purposes. The concept of International) uses MKS measurement units. The
line-force representing the magnetic eld and the total magnetic eld of the Earth is about half an
convention used in force direction relative to the Oersted.
magnetic poles are shown in Fig. 1.4. In petroleum exploration, aeromagnetic sur-
Magnetic surveying, basically depends on the veying are usually conducted to delineate major
variation in the body-ability of being magnetized structural changes of areas such as sedimentary
when exposed to a magnetic eld. This property basins, and regional geological changes includ-
(called the magnetic susceptibility), differs with ing mapping of sedimentary basins and major rift
different materials. Sedimentary rocks are gen- zones. On smaller scales, magnetic surveying can
erally of small susceptibility values compared be used to detect magnetic minerals such as
with metamorphic or igneous rocks. For this magnetite and other iron ore deposits. The
1.3 Geophysical Exploration Methods 7

L L L

Fig. 1.5 Principle of the Wenner conguration used in electrical resistivity surveying. I and V denote electric current
and voltage respectively. L is the distance between electrodes

method is used to explore near-surface geological Articial-Source Electrical Methods


changes (such as dykes and sills). Another typi-
(i) Electrical Resistivity Surveying:
cal eld of application of the magnetic surveying
In the electrical resistivity method, a direct
is exploration of buried archaeological objects.
electric current (DC) electric, or very low fre-
As it is the case with all of the other explo-
quency current (AC), is introduced into the
ration potential methods, the magnetic method
ground. From the voltage drop measured over a
suffers from the ambiguity problem in the inter-
dened distance, the effective (apparent) resis-
pretation process. The degree of uncertainty due
tivity is mapped and then interpreted in terms of
to this problem is reduced by using additional
geological changes. The simplest eld technique,
independent geological information.
called Wenner conguration, consists of two
current electrodes and two voltage measuring
1.3.4 The Electical Method electrodes as shown in Fig. 1.5.
The two approaches (forward and inverse)
This method is based on the relationship between modeling processes are applied in interpreting
the electrical conductivity (or electrical resistiv- resistivity data. The more widely used method is
ity) property and electrical current (or electrical done by applying a trial-and-error method
voltage). Some electrical surveys are based on (model-analysis method). By comparing of the
articially-generated electrical currents while observed data with curves computed for dened
others are using the Earth naturally-generated geological models, a subsurface geological
electrical currents. These are summarized in the model is determined.
following table: Resistivity surveying is particularly effective
in exploring discontinuities as in determination
By articial electrical eld By natural electrical of layering, faults, sills and dykes, especially
methods eld methods when these are not too deep features. The method
(i) Electrical Resistivity (i) Self Potential (SP) is usually applied in engineering geophysical
(ii) Induced Polarization (IP) (ii) Telluric Currents
studies, as in determination of the water table and
other ground-water investigations.
(iii) Equipotential & (iii) Magneto telluric
Electromagnetic (EM) Currents (ii) Induced Polarization Surveying
The Induced Polarization method (IP method)
Electrical methods are mainly used for uses the decay time of an electric potential,
exploration of ore-bodies, minerals, ground water induced after an electric current fed to the ground,
resources and for relatively shallow geological is switched off. The phenomenon is associated
anomalies. Because of lack of deep penetration, with electrochemical reactions activated by the
its use for oil exploration is limited. Most of the electric current which disturbs the ion distribution
methods are effective only for shallow depths in the affected ground material. The voltage level,
which are not exceeding few hundred meters. attained due to the applied electric current, gets
8 1 Introduction

a bird trailed behind. The EM method is used


for exploring base-metals and ground water
voltage

accumulations.
Natural-Source Electrical Methods
time, t There is another less commonly applied electrical
t0
group of methods which are dependent on mea-
Fig. 1.6 The procedure used in the IP method. Voltage surements of the earth natural electric eld. Self
drop after the applied electric current is switched off at Potential (SP), Telluric Current and Magneto-
time (t0)
Telluric methods are typical examples of such
techniques. These methods are used to determine
back to normal state after the applied electric large-scale crustal structural variations as sedi-
current is suddenly stopped at a certain instant. The mentary basins and regional-scale geological
process is sketched in Fig. 1.6. variations.
The method is used mainly for purposes of (i) Self Potential (SP) Surveying
metallic minerals and ground water exploration.
Mineralized zones found in the upper part of the
For the eld surveying, the same electrode set up
Earth crust and some ore bodies found at shallow
for resistivity is also used. Thus using four depths, develop their own natural electric elds.
electrodes (called dipole-dipole array) is used in
In particular, a metallic sulde body which has
IP surveying. The polar dipole system is pre-
part of it above the water table and the rest of it, is
ferred where the voltage electrode spacing is below it. With this condition, such a phenomenon
decreased to minimize the effect of wire-currents.
is created. The body acts as a natural battery
(iii) Equipotential and Electromagnetic where ions, due to differential oxidation, move
(EM) Surveying from one part of the body to the other. The
Equipotential surveing uses the electric potential electrical currents are formed as a result of
eld generated by a xed electric-current electrode chemical reaction that takes place with the aid of
while the electric potential is mapped by a moving the electrolytes present in the host medium.
electrode. Instead of using direct current The potential difference measured on the sur-
(DC) electrodes, another method, the Electro- face will map the potential anomaly caused by the
magnetic (EM) method, uses an electric alternating body. The behavior of the electric potential due to
current (AC) of few hundred-to-few kilohertz, presence of a sulde body is shown in Fig. 1.7.
induced into the ground by a source-coil and
received by another coil, the receiver-coil. The electrical
transmitter coil induces an alternating (primary) potential
magnetic eld, while the receiver coil senses the +
ground-generated alternating (secondary) mag- 0
netic eld combined with the primary. The com-
bined magnetic elds induce AC in the secondary -
coil. This current is then measured and converted ground level
into the combined magnetic eld at the location of
oxidization
the receiver. From comparison of the combined region
magnetic eld with the primary magnetic eld + water table
S
(which is known), the contribution of the anoma- -
lous body is determined and then interpreted into
current
geological information. flow lines
In addition to ground surveying, there is the
airborne EM surveying where the transmitter coil Fig. 1.7 Sulde body (S) acting as a natural battery,
is xed to the aircraft and receiver is mounted in causing abnormality in the generated electrical potential
1.3 Geophysical Exploration Methods 9

The SP method is typically used to map the summarized in the following table (Maton et al.
electrochemical potential variations, generated 1995, p. 269).
by buried sulde bodies.
Radiation Radiation Energy
(ii) Telluric Currents Surveying type characteristics characteristics
The term (telluric currents) is used for the natural
Alpha Positively-charged Can be stopped
electric currents that flow horizontally in the particles protons by a sheet of
upper part of the Earth crust. Variation of the paper
current-density, over the earth surface is gov- Beta Negatively-charged Can pass
erned by the rock-resistivity changes. Thus, if a particles electrons through as
salt dome, for example, is found imbedded much as 3 mm
of Aluminum
within relatively high conductivity-formations,
the lines of currents flow will by-pass the salt Gamma High-frequency Can pass
rays electromagnetic through several
body causing distortions in the potential gradient waves centimeters of
in the overlying surface cover. From the detected Lead
anomalous potential gradient, the geological
changes causing the changes in the potential All of the three types of radiation have the
gradient, can be determined. ionizing ability when colliding with atoms. This
(iii) Magneto-Telluric Currents Surveying effect is the basis of radioactivity detection.
Another related method, the so-called A detection instrument (the Geiger counter)
(magneto-telluric method), involves simultane- was designed by Hans Geiger in 1928. It consists
ous measurements of both voltage created by the of a tube lled with an inert gas (helium or
telluric currents and the magnetic eld induced argon) at a reduced pressure. When radiation
by these currents. Plots of the alternating voltage enters the tube, it removes electrons from the
and that of the associated magnetic eld (as atoms of the gas, making them positively
function of frequency) can give information on charged ions. The electrons move to wire in the
the resistivity as function of depth. tube, setting up an electric current which is
amplied and fed into the recording device. The
(iv) Audio-Frequency Magnetic Surveying current produces clicking sound that pulsates at a
There is still another closely related method rate which is depending on the radiation strength.
which uses the audio-frequency variations in the A counting unit (counter), attached to the wire,
Earth electric eld, to study the earth resistivity measures the created current and gives a pro-
variations. This method is called Audio- portional radioactivity-reading.
Frequency Magnetic method, termed (AFMAG)
method.

1.4 Oil Well Drilling


1.3.5 The Radioactivity Method
Drilling of the oil exploration well is considered
Some naturally occurring substances, such as to be the last phase of any oil exploration project.
Uranium and Thorium, emit particles and radia- This type of work provides the direct information
tion energy, as a result of atomic disintegration. as regards geological and geophysical properties
These chemical elements are called (radioactive of the subsurface rock medium. The obtained
elements), and the phenomenon of the radiation, drilling data give direct recognition of the rock
emitted in the form of particles and rays, is called formations and their mineral and fluid contents
(radioactivity). Three types of radiation (Alpha-, (hydrocarbons and water). Associated with the
Beta-, and Gamma-radiation) are released in a drilling process, are certain activities which help
radioavtivity process. Radiation and energy in getting these useful data. Most important of
characteristics of these types of radiation are these are laboratory analyses of the rock cuttings
10 1 Introduction

especially drilled for injecting water or gas to


enhance the reservoir pressure.
An exploration well provides the following
useful information:
Drilling parameters (like drilling rate) give
indications on the physical nature of rocks
penetrated by the drilling process.
Rock cuttings and cores provide direct
information on the rock lithology and thick-
nesses of the penetrated rock formations.
Circulating drilling mud, cores, and cuttings
provide indications of presence of
hydrocarbons.
Well logs (as electrical logs, radioactivity
logs, and sonic logs) furnish valuable infor-
mation on geological changes and on hydro-
Fig. 1.8 A typical rotary drilling rig, set up for oil-well carbon contents.
drilling operation

1.4.2 The Oil-Well Rotary Drilling


and from well cores. Also, drilling parameters
and the well geophysical logging give valuable The common technique applied in drilling oil
geological and geophysical information. wells is what is called (Rotary Drilling). The
The common way followed in drilling is what drilling process is done by rotating a column of
is known the rotary drilling which is done by use steel pipes, with a drilling head (called the Bit)
of the drilling rig, especially designed to carry attached to its end. The rotating bit carves a hole
out this operation. A photo of typical oil-well rig through the rocks and in this way a drill-hole is
is shown in Fig. 1.8. made. Drilling mud is pumped through the pipe,
returning to surface via the space between the
drill pipe and well wall. Mud circulation will
1.4.1 Drilling of the Exploration-Well carry rock cuttings to surface and provide cool-
ing and lubrication to the drilling process.
Since the rst 69-ft oil-well (drilled by Drake in Drilling operation is done by the drilling rig,
Pennsylvania, USA in 1859), drilling technology which is a mechanical steel structure designed in
has witnessed great advances in the capability of the form of a tower of height of (120150 ft).
drilling deep wells and in the control on the A schematically representation of a typical
direction of the drilled well. Inclined wells and oil-well drilling rig, with its principal parts, are
even horizontal drilling are now in common shown in Fig. 1.9.
application. The drilling rig, shown in Fig. 1.9, is made up
Drilling of the rst oil well after completing of the drill tower (called derrick), drill pipes,
the geophysical exploration activities is called drilling floor (Kelly Bush), and other facilities to
(Exploration drilling) or (Wildcat drilling). There provide energy for rotating the pipe column and
are other types of drilling labeled according to to circulate the drilling mud. The grinding pro-
the purpose of the drilled well. For instance, it is cess is executed by the rotating bit attached to the
called Production drilling when the well is end of the drill pipe. The more common types of
designed for normal oil production, and obser- bits are the grinding bits which produce rock
vation drilling for monitoring the behavior of the cuttings that are removed by the circulating mud,
oil reservoir. Wells for water or gas injection are and the coring tubular bits producing rock-cores.
1.4 Oil Well Drilling 11

crown block Flushed Zone (FZ)


Natural fluid is completely replaced by the dril-
ling fluid (several inches).
travelling block
derrick Invasion Zone (IZ)
Drilling fluid inltrated into region surrounding
drill pipes mud hose the well (few inches as in shale to several feet as
in porous sandstone).
After removing of the drill pipes, a pipe col-
power unit
pump umn is lowered into well, then cement is pumped
in, to hold it tight in place. Casing prevents
kelly bush rotary table mud tank
caving and fluid seeping. However, to allow oil
and water to flow into well, the casing is later-on
kelly perforated at the appropriate places.
Electric logs cannot usually be run with cas-
ing and sonic logging is severely disturbed. In
drill pipe
general all logging processes are carried out in
uncased (open-hole) wells.
drilling bit

Fig. 1.9 A simplied sketch showing the main parts of a 1.5 Well Geophysical Logging
rotary drilling rig

After completion of the drilling operations of a


1.4.3 The Drill-Hole and Well Casing
well, a group of technical activities (well log-
ging) are done to extract direct information on
During the drilling process, drilling mud is cir-
the rock formations penetrated by the drilling
culating carrying with it the produced rock cut-
process. Study of the penetrated rock formations
tings. According to the extent of penetration into
(commonly referred to as formation evaluation)
the well wall of drill-fluid, three types of zones
includes examination of its contents of rock
are recognized. These are Fig. 1.10:
cuttings, detection of hydrocarbon matter, and
Mud-Cake Zone (MZ) documentation of lithological and palaeontolog-
This is a thin coating of well-wall which is of few ical changes. The greater bulk of activities done
tenths of an inch. at this stage is the geophysical logging of the
drilled well which is always done as a concluding
stage of the drilling operation of drilled explo-
ration wells. Well logging gives direct determi-
MZ MZ nation of petrophysical properties of the
FZ FZ
penetrated rock formations. These are essential
practices needed in the rock-formation evaluation
undisturbed IZ IZ undisturbed
formation formation and reservoir characterization processes.
drill
hole The well log (or wireline log, as it is often
referred to) is recorded by a special well logging
tool (called the sonde), carrying sensors which are
lowered into the hole by a cable. The standard
procedure is to start the measurements at the bot-
tom of the hole and moved upwards through the
Fig. 1.10 Mud invasion-zones into the well wall,
labeled: mud-cake zone (MZ), flushed zone (FZ), and borehole. Measurement of a certain geophysical
invasion zone (IZ) parameter is done either continuously or at discrete
12 1 Introduction

the potential drop (V) is related to the resistivity


(q) as follows:
recorder From Ohms Law we have, the potential
(V) across the radius (r) of a sphere is expressed
as (V = I R = I q r/4p r2 = I q/4p r), R is the
electrical resistance.
log
sonde
r2
well r1

Fig. 1.11 Principles of well logging


Thus, over the separation distance (r1 r2),
points. The output is recorded and normally pro- we get:
duced as charts showing the value of the measured
parameter as function of well-depth (Fig. 1.11). DV Iq=4p1=r1 1=r2 Iq=4p r1
The well logging process results in for r2 r1 :
measurements-data normally plotted as charts
called (well logs). Since it was introduced (by Hence, resistivity (q) is given by:
Schlumberger in 1928), the geophysical well
logging technique has vastly developed into an q 4pr1 DV=I
indispensable formation-evaluation tool for the
rocks penetrated by drill-holes. Logging methods There are two types of congurations, the
can be divided into the three principal methods: normal-sonde and the lateral-sonde logging
electrical-, radioactivity-, and acoustic-logging. (Fig. 1.12).
The resistivity log expresses measurements of
the electrical resistivity in the usual (ohm units) for
the rock medium surrounding the drill-hole. Pene-
1.5.1 Electrical Logging
tration distance of the electric eld in the penetrated
rocks depends on the electrode spacing-distance.
Basically, electrical logging involves measure-
The larger the electrode spacing, the greater is the
ments of the variations of electrical resistivity and
penetration distance (Fig. 1.13).
natural potential of rocks down the drilled well.
The normal sonde electrode-conguration
Depending on the applied electrode conguration,
includes two electrodes: one source and one
the following techniques are in common use.
receiver spaced few feet apart. The instrument
(i) Electrical Resistivity Logging reading which is given in apparent resistivity
This type of electrical logging is based on a con- units reflects the properties of the region near the
guration whereby a DC (or low frequency AC) source-electrode. The other type (the lateral-
source electrode is lowered down the drill-hole and sonde conguration) is equipped with three
measuring the potential drop across a set of electrodes; one source and two receivers. A vari-
potential electrodes. The output is a continuous ation was made on the lateral sonde whereby the
record of the variation of the electrical potential (or current rays spread out horizontally (focused
the corresponding apparent resistivity), with depth. rays) rather than radially (unfocussed) spreading.
This method must be applied in uncased wells. This modied procedure is called (laterologging)
Assuming spherical radiation of electric cur- and the produced resistivity chart is called
rent is emitted from the point-source electrode, (laterolog). The lateral logs give more accurate
1.5 Well Geophysical Logging 13

Fig. 1.12 The two types


of electrode congurations I I V
V
used in resistivity well
logging: normal sonde and
lateral sonde congurations

S
r1

r2

normal-sonde lateral-sonde
configuration configuratio

eld which induces electric currents into the rock


formations. These currents, in turn, create sec-
ondary magnetic eld which induces, in the
receiving secondary coil, an AC which varies
s s s
with the resistivity of the formations. The
coil-sondes conguration is shown in Fig. 1.14.
Like lateral logging, there is a variation made
d d d on induction logging to give focused current
radiation, for getting sharper boundary indica-
Fig. 1.13 Lateral distance (d) penetrated in resistivity tion. Induction logging is used in wells lled
logging is proportional to electrode spacing (s) with conducting, or no-conducting, drilling-mud.
(iii) Spontaneous Potential (SP) Logging
and sharper boundary detection, in addition to This logging method depends on measurements
reduction of effects of the mud and hole-diameter of the natural electric potential (in millivolt units)
variations. of the rock medium surrounding the surveyed
Resistivity logs help in the diagnosis of types well. It is normally referred to as self-potential or
and boundaries of formations. For example, low spontaneous potential (SP) logging. It is used
resistivity indicates higher porosity and perme-
ability of water saturated formations, whereas
increase of resistivity can reflect existence of oil AC
recorder
generator
and gas. In general, low-porosity rocks (as shale)
and porous rocks saturated with salty water
exhibit low resistivity. On the other hand, porous
rocks saturated with low-salt fresh water, or ground current
primary coil
saturated with oil will exhibit high resistivity. On
this basis, resistivity logs serve as important and
effective indicators for presence of oil.
secondary coil ground current
(ii) Electrical Induction Logging
The logging sonde uses coils instead of normal
metal-piece electrodes. A primary coil carrying Fig. 1.14 Coil-sondes conguration used in induction
an AC current creates an alternating magnetic logging
14 1 Introduction

Fig. 1.15 Conguration


used in spontaneous V
potential logging with - millivolt +
schematic representation of
the SP log
shale line
sand line

sonde

SP-Log

only in an uncased holes lled with conductive be drawn for sandstone SP-value. The inflection
mud. The measuring sonde is of simple congu- points in the SP-log indicate formation
ration. It consists of only two electrodes: one is boundaries.
lowered into the well by an insulated cable and
the other is xed at the ground surface
(Fig. 1.15). 1.5.2 Radioactivity Logging
This method, which needs no articial current,
can detect natural potential differences which Radioactivity is the phenomenon of emission of
develop at formation boundaries. Compared with particles and photons of electromagnetic energy
resistivity logs, the SP logs give more sharp from an atom. This radiation process occurs
changes and hence more accurate formation- either naturally from unstable nuclii or induced
boundaries. Like lateral logging mentioned by bombarding of stable nuclii by photons or
above, there is a variation made on induction atom particles. Examples of such radioactive
logging to give focused current radiation for elements are Uranium, Thorium Rubidium and
getting sharper boundary indication. Potassium 40 which is most commonly found in
Interpretation of the SP logs depends on the shale and clay and less in sandstone and
manner of the electrical potential variation. As a limestone.
general rule, changing of the log values towards There are three types of radiation: Alfa parti-
positive potentials is considered as an indicative cles formed of charged helium nuclii, Beta par-
of impermeable rocks as shale or tight limestone ticles formed of high speed electrons, and
or tight sandstones. When the change is towards Gamma rays of electromagnetic wave-energy.
negative potential, it is interpreted as being due Out of these, only gamma ray is used in well
to porous sandstones or porous limestone. In this radioactivity logging. The detection instrument is
way, the SP logs give useful indications on a Geiger or more usually is the scintillation
lithology, water salinity and help in determina- counter which consists of a special crystal (like
tion of formation boundaries. In particular, it sodium iodide), which emits flashes of light as
shows the boundaries of formations in sand-shale they absorb gamma-ray photons, hence the name
sequences. These are determined with the help of (scintillation counter). A photoelectric tube con-
drawing lines (sand and shale base-lines) in the verts these flashes into electric currents which are
produced SP log (shown in Fig. 1.15). It is displayed in the form of a continuous chart (the
commonly observed that shale beds give the radioactivity log).
same level of SP-readings allowing for drawing a Radioactivity logs provide important infor-
straight line indicating shale SP-value. This is mation on rock lithological types, especially on
called (the shale line). Similarly a (sand line) can those containing certain concentrations of
1.5 Well Geophysical Logging 15

radioactive minerals. Thus, these methods are geological


ideal indicators of shale and clays which, by column

nature, contain radioactive minerals in their log


makeup.
Three methods of gamma radioactivity log-
ging are in common use. These are: Natural
Gamma-ray, Gamma-ray rock-density, and neu- D1
tron Gamma-ray Logging methods. D2
(i) Natural Gamma-Ray Logging S
This method uses a detector mounted on a sonde
to measure the naturally emitted gamma rays mud cake
from the radioactive minerals existing in the rock
Fig. 1.16 Principle of gammagamma log recording.
formations. Unlike electrical logging, gamma-ray
Detectors, (D1) and (D2) record secondary and mud-cake
logging can be run in cased wells, as well as in radiations respectively and the symbol (S) represents the
uncased wells, with detection penetration of few source. Logs (q) and (Dq) represent bulk density and
feet from the well walls. The resolution power of mud-cake correction respectively
formation-boundaries is affected by the counting
time of the instrument and sonde logging speed. rate and the back-scattered (secondary radiation)
Reasonable results are obtained with a counting are proportional to electron density which is, in
time of 2 s and sonde speed of 15 cm/s. Mea- turn, proportional to rock-formation density. The
surements can be made in cased wells, but the principle of the gamma-gamma logging is shown
intensity of radiation is reduced by about 30 % in in Fig. 1.16.
this case (Kearey and Brook 2002, p. 244). In practice, the gamma-ray detector is shiel-
In general, log values are interpreted as increase ded to record only the secondary radiation, and
of the shale percentage, while the fall in the the sonde is rmly pressed to well-wall and
log-values is interpreted as indication for sand- moved slowly (less than 30 ft/min) in order to
stones and limestone rocks. To aid interpretation, maintain good contact. The produced chart
it is possible to draw shale-lines and sand-lines on (gamma-gamma log) expresses the formation-
the log chart. An advantage of the gamma log is its density log.
capability of differentiating between shale and A variation to the method is introduced to
sandstones independent of the porosity and per- provide corrections for mud-cake effect. As
meability characteristics of the rocks. shown in the gure, above, a secondary detector
(ii) Gamma-Ray Rock-Density Logging D2 is included in the sonde which is responding
The sonde contains a gamma-ray source and a to mud-cake and small wall irregularities. The
scintillation counter to detect the gamma-ray resulting log (called Compensated Density Log)
which is back-scattered from the formations and shows both of the bulk density (q) and the
received by a detector xed at a certain distance density-correction log (Dq).
from the source. It is also called (gamma-gamma Interpretation of the density log is based on
logging) method. the direct proportionality existing between the
The gamma-ray photons collide with the recorded scattered gamma ray intensity and the
electrons of the elements in the formation number of electrons found in the scattering rocks.
resulting in loss of photon energy and back The number of the scattered electrons is, in turn,
scattering of gamma-ray which has a wavelength proportional to rock bulk density.
different from that of natural gamma-rays. This is This type of log can be used in computing
called (Compton Scattering Effect). The collision porosity () by using the following relationship:
16 1 Introduction

displayed in the appropriate measuring units or in


porosity percentages directly.
The intensity level of measured radiation is
proportional to concentration of hydrogen, which
Gamma exist in water, in hydrocarbon, and in hydrous
radiation minerals such as silicate-clays, micas, amphiboles,
Hydrogen
H nucleus and gypsum. Thus in carbonates and sandstones,
N hydrogen source is water and hydrocarbon found
neutron source in the pores of the rock. In shale, however, mica
N and clay minerals contribute to hydrogen content
Pressing device as well as from pore water. In this case other types
N of logs (e.g. gamma-ray logs) are needed to dis-
neutron tinguish shale from water-saturated porous sand-
stones or limestones. In general neutron logs are
best in following up the porosity variation of
Fig. 1.17 Principle of Neutron Gamma-Ray logging
porous rocks using the direct proportionality
between porosity and gamma-ray intensity-level.
In short, Neutron logs, depend on the
gamma-ray generated from neutron bombard-
qm ql =qm qf
ments of hydrogen atoms. The generated gamma
where, (qm), (ql), and (qf) represent matrix den- ray, which is proportional to concentration of the
sity, log read-density, and pore fluid density hydrogen element in the penetrated rocks, is
respectively. dependent on the water and hydrocarbon fluids in
the rocks. This also means that high log values
(iii) Neutron Gamma-Ray Logging indicate high porosity. It is useful to note that in
The sonde consists of a neutron source and case of presence of hydrocarbon gas in high
detecting scintillation counter placed at a xed porosity formations, density logs (gamma-ray
distance apart. The source is a small radioactive rock density logs) is expected to give low log
body (such as Plutonium-Beryllium) which emits values compared with the neutron log vales at the
neutrons during the process of radioactive decay. same zone, where the neutron readings expected
When the source-generated neutron collides to be relatively high. This means that both of the
with a hydrogen nucleus (which is of a matching logs (density and neutron logs) are necessary in
mass) its kinetic energy is reduced to an extent order to detect hydrocarbons (oil or gas).
that it can be captured by a large nucleus (as the
hydrogen nucleus) causing emission of a sec-
ondary gamma radiation (capture gamma radia-
tion). The principle of the logging is sketched in
1.5.3 Acoustic Logging
Fig. 1.17.
In the logging process, the sonde is moved at The purpose of this type of logging is basically
for getting information on velocity of propagating
a speed of 30 ft/min with 2 s for the counting
acoustic (seismic) waves. Sonic logging, well
time. A skid (pressing device) is provided to
keep the sonde in close contact to the wall. The velocity surveying (well shooting) and VSP sur-
veying are included in this type of well-logging.
gamma radiation, generated by the capturing-
phenomenon, comes from the material sur- (i) Sonic Logging
rounding the drill-hole. The produced log (which The logging sonde, in its standard form, consists
can be run in a cased or non cased well) is of two receivers about 1 ft apart and a source at
1.5 Well Geophysical Logging 17

double-source sonde single-source sonde


transit time, sec
geological column 140 90 40

shale

S S
limestone
R R

R R sandstone
S
shale

borehole compensated none-compensated

Fig. 1.18 Conguration of source (S) and receiver (R) of


Fig. 1.19 Sketch sonic log of a hypothetical geological
the sonde employed in sonic logging
column

about 3 ft from the nearest receiver. To correct Sonic logs provide the interval transit time
for tilting and hole-irregularities effects, a dual which represent travel-time (usually in
source sonde is used, making what is called a microsecond units) of the P-wave in covering a
borehole-compensated sonde (Fig. 1.18). distance (usually 1 ft). This parameter (the
The electronic structure of the sonde is interval transit time) is useful in computing the
designed in such a way, that the output is made porosity () by using the following relationship:
to be the difference in the travel-times to the two
receivers. The time difference, measured in Dtl Dtm =Dtf Dtm
time-units per 1-ft, called (interval transit time),
is plotted (normally in micro seconds) against where, (Dtm), (Dtf), and (Dtl), represent transit
depth to give the continuous wiggly curve known times of matrix material, of pore fluid, and log
as the (sonic log). In the compensated sonic read-transit time respectively.
logging, seismic pulses are emitted alternately
(ii) Well Velocity Surveying
from the two sources and the transit times from
A hydrophone-type detector is lowered down the
the two oppositely traveling refracted P-waves
well which is lled with the drilling fluid. The
are averaged electronically. The output (transit
travel time of the seismic wave generated by a
time) is plotted against depth, giving the
surface-placed shot and received by a detector
borehole-compensated (BHC) sonic log
placed at a formation boundary is recorded. The
(Fig. 1.19).
full log is obtained by repeating the recording at
The borehole compensated sonde (BHC)
each boundary traversed by the well. From the
gives an average interval transit time which is
measured travel time of direct arrivals, the av-
plotted on a paper strip. The produced log in this
erage velocity and interval velocity are plotted as
case is normally referred to as BHC sonic log
function of well-depth. The velocity survey is
which is used to identify lithologies, determine
also called (check-shot survey). Principle of the
formation boundaries, and in computing syn-
velocity survey setup and velocity-depth plot is
thetic seismograms. The interval transit time can
shown in Fig. 1.20.
be integrated down the well to give the total
The obtained velocity information is used to
travel time. This type of logging can only be run
calibrate the sonic log and check the sonic-log
in an open (uncased) hole.
18 1 Introduction

Fig. 1.20 Principle of the


well velocity surveying and seismic source average velocity
a typical velocity-depth
plot obtained from the
survey

seismic
detector

depth

integrated time (hence the name, check-shot normally applied to refer to the direct and
surveying). reflected waves respectively. The ray paths of the
VSP shooting and the corresponding travel-time
(iii) Vertical Seismic Proling (VSP) plot of the recorded waves are schematically
Basically the measurement set-up is the same as shown in Fig. 1.21.
the check-shot recording system. The difference A seismic pulse is generated on the surface
is in the recording duration time which is here from dynamite explosion or from an air gun
extended to allow recording reflected waves as submerged in a water lled hole. This is repeated
well as the direct arrivals. At each detector at all of the hydrophone positions, normally
stop-location (normally at 25 m spacing), the positioned at 25 m-spacing down the well.
recorded seismic trace is allowed to include With an appropriate processing sequence, the
events from up going reflected waves in addition nal corrected VSP section is produced. Process-
to the rst arrival (direct wave) event. The terms ing includes, data editing, correction to vertical
(down-going wave) and (up-going wave) are time and velocity ltering for separating unwanted

Fig. 1.21 Ray-path seismic


diagram and the source
corresponding travel time
time
plot of a VSP survey of a
depth
geological model made up
of three reflectors R1 R2 R3

R1
down- tube up-going wave
R2 going wave
wave
R3
Ray-paths of Travel-time plot
direct and reflected waves of the tube wave,
from a three-reflector model down-going & up-going waves
1.5 Well Geophysical Logging 19

events. It is evident from the VSP section that Drilling information (drilling rate, mud den-
downward events (primary and multiples) increase sity variation)
in time with depth, and the upward events (primary Caliper and dip-meter logs.
and multiples) decrease in time with increasing
The main information which can be obtained
depth. By arranging the produced seismic traces
from well logs can be summarized as in the
(25 m spaced down the hole) a seismic VSP sec-
following table
tion is obtained. Each seismic trace of a VSP
section contains events from down-going waves
Log Type Application Comment
(direct and multiples arrival) and from up-going
Electrical Fluid-type Logging is
waves (reflections and multiples). Logs identication done in
The main application of VSP is providing Resistivity Porosity uncased
seismic section of reflectors which have not been Induction evaluation wells
reached by the drilling. In fact the VSP data are Spontaneous Boundary
Potential determination
equivalent to both the check-shot data and the Shale and sand
sonic-derived synthetic seismogram. lines
(iv) Other Logging Techniques Radioactivity Shale/sandstone Logging is
Logs Formation done in
Some logs are used to give geometrical infor-
Natural density cased wells
mation on the deviation from the vertical gamma-ray Porosity
(hole-drift angle) as well as the azimuth of the Gamma-ray Fluid type
deviation. Other logs are made to give the bed- density
Neutron
ding dip (dip-meter logs) and measurements of
gamma-ray
the well-hole diameter (caliper logs).
Acoustic Lithology types Logging is
On more limited scale of application, other Logs Formation done in
types of well logging have been used. Specially Sonic logs Boundaries cased and in
modied borehole-gravimeters, magnetometers, Well Synthetic uncased
and thermometers are examples of such tools. velocity seismograms wells
VSP Seismic-velocity
Geothermal prospecting is applied to detect functions
geological features which affect heat flow such as Reflection
shallow salt domes, faults and dykes. Also identication
ground water is investigated by this method.

1.5.4 Log Interpretation 1.6 Latest Developments in Well


Logging
The process of analysis and interpretation of well
logs is sometimes called (Formation Evaluation). Since the introduction of well logging by Sch-
Normally all the logs obtained for a well and lumberger brothers in 1927, logging techniques
their analysis-results are presented in one com- passed through a series of advances which are
bined display called the (composite log). This concerning measurements approach and mea-
comprehensive log usually contains all (or most) surement accuracy. The new techniques, aided
of the following log data: with the specialized computer software, have led
Geological column showing lithological and to increased work efciency and increased
palaeontological information detection resolution. The two prominent devel-
Borehole compensated sonic log opments happened to the logging techniques are
Gamma-ray and Neutron logs the logging while drilling (LWD) procedure and
Resistivity and SP logs borehole imaging tools.
20 1 Introduction

1.6.1 Logging While Drilling running-in and cementing the casing is consid-
Technique ered as integral part of the completion process.
To prevent caving in of the well walls, drilling
In late 1980s a new logging procedure, logging is stopped at various depths. and a steel pipe
while drilling (LWD), was introduced. This (casing) is lowered into the well and xed in
technique is similar to the conventional wireline position by pumping cement mixture in the space
logging except that it (the LWD) uses sensors between the casing pipes and the well walls
which are incorporated in the drilling bit (cementing process). According to the planned
assembly, and thus the measurements are made drilling program, drilling after casing being
during the drilling process. This measurement cemented is continued with a different (usually
procedure coupled with specialized computer smaller) bit-size. At this stage, the drilled hole is
software provides fast geological information as lined by the appropriate casing which is likewise
lithology, porosity, fluid contents, and drill-hole cemented in. The drill-casing-and-cementing
direction. phases are repeated until the nal planned total
depth is reached. The end result of the casing
operations is a set of concentric pipes each of
1.6.2 Borehole Imaging Tools which is ending at the earth surface. When the
planned total depth is reached, all used drill pipes
Borehole imaging is a logging technique that can including the attached drilling bit, are withdrawn.
provide small-scale images of the borehole wall The next operation is perforation of the parts of
and the penetrated rock lithology. One of these the casing facing the oil bearing formations. The
methods (based on electrical micro-imaging so-produced holes are made to allow the oil to
processes) was developed by Schlumberger
under the trade-name; formation micro-imaging
(FMI). In 1986, Schlumberger developed a
dip-meter called the formation micro-scanner christmas
(FMS). Later versions of this logging tool were tree
called formation micro-imager (FMI) which is
based on micro-resistivity measurements.
The operation of the FMI tool is based on
recording changes (and not absolute values) in
electrical resistivity of rock formations. It pro-
vides, with high resolution power, identication
20-inch casing
of sedimentary characteristics and fracturing
picture in the penetrated rock column, in addition
to measuring formation dip. 13 -inch casing

1.7 Well Completion 5 -inch casing

Well completion in petroleum production, is the


process of making a drilled well ready for pro-
duction (or fluid injection). This principally
involves preparing the bottom of the hole to the
required specications, running in the central 2 -inch tubing
production tubing and its associated down-hole
Fig. 1.22 The nal setup of the drilled well after casing
tools as well as perforating and stimulating as and installment of the well head monitoring gauges (the
required. Sometimes, the process of pipes Christmas tree)
1.7 Well Completion 21

seep inside the central production tubing of the equipment in addition to the mechanical system
well. that provide the blowout prevention (BOP) is
The last stage in preparing the well for the called the well head which is also equipped with
production process is xing of a group of flanges an assembly of valves that controls flow, and any
which provide a sealing structure connecting other possible interventions. This assembly is
together all the surface ends of the casing tubes. normally referred to as the Christmas tree
This part of the well which contains pressure (Fig. 1.22).
Seismic Waves
2

2.1 The Fundamental Conditions motion of elastic waves are generated and can be
recorded by the appropriate detection instru-
An application of an external force, on part of a ments. These are normally referred to as the
medium (elastic medium), leads to creation of (seismic waves).
internal opposing forces which intend to resist
the deformations caused by that external force.
Typical forms of the resulting deformations are 2.2 Theory of Elasticity
changes in volume and/or in shape which are
created at the affected location. In consequence, As it is stated above, the fundamental conditions
the medium will return to its original condition for a seismic eld to be created is that the med-
after the external force is removed. This property ium must possess the elasticity property. Two
of resisting of changes in volume and in shape main concepts are governing the propagation of
and return to original conditions after removal of seismic waves in an elastic medium: the (stress)
the external force is called (elasticity). Provided and the (strain). Stress represents the external
that the changes are small, rock media in nature force applied to the elastic medium, and strain is
are considered to be perfectly elastic in nature. the resulting changes in volume and in shape.
As a result of the elasticity property of media, The relation between stress and strain, for a
the changes (volume and shape changes) oscil- particular medium (perfectly elastic medium),
late about their neutral positions and, at the same gives evaluation expressions for the elasticity
time, propagate away from the energy property of that medium. The stress-strain pro-
source-location. Energy transfer in this manner portionality constants are the elastic coefcients
(motion that leaves out no permanent distortions) which serve as measures of the elasticity of a
is commonly referred to as (wave motion). particular medium.
The fundamental condition for the creation The principal types of changes experienced by
and propagation of seismic waves (seismic eld) a medium due to passage of a seismic wave are
is a source of mechanical energy of impulsive re-distribution of the internal forces (stress
type which is initiated within an elastic medium. changes) and modication of the volume and
The energy source may be natural (as in geometrical shape (strain changes). The theory of
earthquake-generated waves) or articial (as in elasticity deals with analysis of these principal
ring of a dynamite charge). In both cases wave effects and the related physical changes.

Springer International Publishing Switzerland 2017 23


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_2
24 2 Seismic Waves

2.2.1 Stress A stress component may be represented by


(Tab) where a (=x, y, z) stands for the area-set an
In the broad sense, stress is represented by a b (=x, y, z) for the direction of the component.
force (called traction) which is acting on a nite Using this convention, the nine components of
area occupying an arbitrary position within the the stress tensor (shown in Fig. 2.2) may be
medium. However, for more precise denition, written as in Table 2.1.
the stress (T) is dened to be a limiting value of For a stressed body which is in equilibrium
the ratio of a force (F) acting on an elementary (i.e. experiencing no rotation), it can be shown
area (DA) which is diminishing to zero. That is: (see Bullen 1965, p. 10) that, due to the sym-
metry of a stress tensor acting on a body in a state
T lim F=DA of equilibrium, we have Tab = Tba. Applying this
DA!0
property to this stress system, we get reduction in
In general, the stress (T) is a vector that can be the number of the components to a total of six
resolved into components parallel and perpen- components which are independent of each
dicular to the area (DA). The normal component other. That is Table 2.2:
(Tz) is called the normal stress or dilatational or These six components are sufcient to dene
pressure stress as it is sometimes called. The the stress system at a point within a stressed
other two components (Tx and Ty) which are in body. Further simplication is possible if the
the plane of the elementary area, are called tan- three areas of an area-set are chosen such that
gential or shearing stresses (Fig. 2.1). the normals to these areas are coincident with the
The stress system within a body is completely directions of the normal stresses. With this kind
dened if, at each point in that body, the normal of set-up the shearing components will all reduce
stress and the two shearing stresses are all to zero leaving only the components (Txx, Tyy,
determined for three mutually perpendicular Tzz) for the denition of the stress system. In
plane areas. such a case, when the shear components become
It follows, therefore, that nine stress compo- all equal to zero, the components (Txx, Tyy, Tzz)
nents are needed to completely dene the stress at are referred to as the principal stresses.
a given point. This nine-component set constitutes For the simple one-dimensional case, when a
what is known as the (stress tensor) at that point. force (F) is acting uniformly at the cross-sectional
Once the nine components are dened at a certain area (A) of a bar or a metal wire, stress is dened
point (with respect to a given area-set) it is pos- as the force per unit area of the cross sectional
sible, through suitable mathematical transforma- area, (F/A). This stress is called tensile stress when
tion, to determine the stress with respect to any the force is a pull-force and it is called compres-
other area-set dened for that point. sive stress when it is a push-force.
The three mutually perpendicular elementary Units of stress is the same as those used in
areas, called an (area set), and the nine compo- measuring pressure (force per unit area), and in
nents of the stress tensor is shown in Fig. 2.2. the SI unit system the unit is Newton per square

T
Tz

Ty
Tx
A

Fig. 2.1 Stress (T) and its components (Tx ,Ty, Tz) acting on the elementary area (DA)
2.2 Theory of Elasticity 25

The three mutually perpendicular areas (A, B, C)


T
B
A

C
y
z

The three stress components (Tx , Ty , Tz) per each of the three
areas (A, B, C)

Tyz B
A
T
Txz Tzz
T T
Tyx

Tyy
C
Txx Tzx

Txy Tzy

Fig. 2.2 The stress components. The mutually perpendicular planes (A, B, C) making up the area-set and the nine
components of the involved stress tensor

Table 2.1 The complete nine components of the stress tensor


Stress component Area (a) perpendicular to: Area (b) perpendicular to: Area (c) perpendicular to:
along the: x-axis y-axis z-axis
(a) x-axis Txx Tyx Tzx
(b) y-axis Txy Tyy Tzy
(c) z-axis Txz Tyz Tzz

Table 2.2 The six independent components of the stress tensor


Stress component Area (a) perpendicular to: Area (b) perpendicular to: Area (c) perpendicular to:
along the: x-axis y-axis z-axis
(a) x-axis Txx
(b) y-axis Txy Tyy
(c) z-axis Txz Tyz Tzz
26 2 Seismic Waves

meter (N/m2) which is called Pascal, where one In general, when a body is subjected to elastic
Pascal is equal to 1 N/m2. stress, both of its size and shape will change. As
it is mentioned above, the resulting changes
represent elastic strains when each point of the
2.2.2 Strain stressed body experiences a displacement of its
own which is different from the displacements
In reference to Fig. 2.3, let us consider the two experienced by the other points of the body. This
points (P1 and P2) located within an unstressed implies that there are two types of strains, namely
body, where the rst point, P1 is located at (x, y, z) the volume strain and the shape strain
and the second point (P2) at (x + dx, y + dy, (Fig. 2.4).
z + dz). Now, we let this body to deform as a
result of a stress system created within it. If the two
points (P1 and P2) were displaced from their 2.2.3 Common Types of Strain
original positions by equal displacements (D,
say), then it is considered that there is no strain Mathematical analyses of strain show that the
taking place. Strain occurs only when there is total strain of a three dimensional body, depends
variation of displacement of any point, within that on only six different derivatives of displace-
medium, with respect to the others. In the lan- ments. These strain components (eab), can be
guage of mathematics, we say that strain depends written down as follows (Richter 1958, p. 236):
on the derivatives of the displacement-
components with respect to the chosen coordi- exx = Dx / x
nates (x, y, z). The concept is claried in Fig. 2.3. eyy = Dy / y
ezz = Dz / z
exy = ( Dx / y + Dy / x) / 2
exz = ( Dx / z + Dz / x) / 2
P1 P2 eyz = ( Dy / z + Dz / y) / 2
(a)

D These equations represent two groups of strain


D
components of a strained elastic body. The rst
group (exx, eyy, ezz) involve purely translational
P1 P2 displacement resulting in compressional or
(b) dilatational strain. The second group (exy, exz,
eyz) involve purely rotational deformation
D D + dD resulting in shear strain. As it is stated in our
discussion of stress, the compressional (or
dilatational) strains are called (principal strains)
Fig. 2.3 Displacement of two adjacent points in a when the shearing strains are all of zero values.
medium under stress. a Case of equal point- Common types of strains are cases of compres-
displacements giving no-strain state. b Case of different
point-displacements giving the strain state sion, bulk contraction, tension, and shear strains.

volume strain Original shape strain


unstrained
body

Fig. 2.4 The two types of strain; volume and shape strains
2.2 Theory of Elasticity 27

Fig. 2.5 Types of simple (a) (b)


strains (a) and no-strain
changes (b)

compression strain tension strain rigid translation

contraction strain shear strain rigid rotation

Rigid body-translation and rotation represent (or compressional) stress. The longitudinal strain
cases of no strain, since no volume and no shape (e) is dened to be the change in length in a
deformation are involved. In Fig. 2.5, an ele- certain dimension, of a body under stress relative
mentary cube (shown here in plan) is used to to its original length. For a rectangular lamina of
show simple types of elastic deformation (strain) dimensions (Dx by Dy), the longitudinal strains
and no-strain changes. (ex and ey) in the x and y directions are dened as
In general, an elastic body under stress can (Fig. 2.6):
experience two types of distortions; changes in The longitudinal strains can be extensional
volume and changes of shape. These changes, (tensile strain) or compressional (contraction
which occur as result of stress, are expressions of strain). Longitudinal strains (ex and ey) are
the physical properties of the stressed body. In its dened as:
simple form, elastic strain can be divided into
two main types. These are: the volume-changing ex Dx =Dx
strain (leading to body compression or dilatation) ey Dy =Dy
and the shape-changing strain (leading to body
shape distortion). where (Dx and Dy) are the changes in length in
x and y directions respectively.
The minus sign that appeared in the ey
2.2.4 The Volume-Changing Strain expression is entered to denote that the change
(Dy) is compression which is in opposite direc-
The familiar example on this type of strain is the tion to the dilatation change (Dx), in the x-
longitudinal strain of a body under an extensional direction.

Dy
y y

x x Dx
ex = Dx / x ey = - Dy / y

Fig. 2.6 Denition of the longitudinal strain as applied for a rectangular lamina of dimensions (Dx by Dy)
28 2 Seismic Waves

Fig. 2.7 Concept of shear


or angular strain, exy where:
(exy a b=2
Dx =Dy Dy =Dx=2
y y

x x

Fig. 2.8 Pure rotation and


pure translation changes of
a body, are not considered
to be elastic strains

pure body rotation pure body translation

2.2.5 The Shape-Changing Strain displacements. When the displacements are equal
a body may experience pure translation (rigid
As longitudinal strain gives expression for the body-translation) or pure rotation (rigid
volume changes resulting from stress application, body-rotation), as shown in Fig. 2.8.
the shear strain gives the corresponding measure Rigid-body changes which do not involve
for the shape deformation. Using the example volume or shape changes, such as these, are not
above (Dx by Dy rectangular lamina). The shear considered to be elastic strains.
strain (also called angular strain) is considered to
be the average of the two angles by which two
neighboring sides rotate as a result of the shearing 2.2.6 The Cubical Dilatation
stress. Thus, the shear strain (exy) is dened as:
exy / b=2 A parameter, closely related to longitudinal strain
and of special importance in the theory of elas-
where (/ and b) represent the angles of rotation ticity is the Cubical Dilatation (h). At a certain
of the two sides (Dx and Dy) brought about by point within a strained medium, this is dened as
the shear stress (Fig. 2.7). the fractional change in a unit volume surrounding
Since these two angles are very small (usually that point. Thus, for a three dimensional body with
so, in seismic-eld conditions), they can be rep- longitudinal strains (exx, eyy, ezz), the cubical
resented by their corresponding tangents, giving dilatation can be computed as follows:

exy a b=2 Dx =Dy Dy =Dx=2 h 1 exx 1 eyy 1 ezz  1

where the angles (a & b) are in radians. For small strains exx, eyy, ezz (which is the
It should be emphasized that strain occurs case in seismic-eld conditions), the products of
only if the body particles experience unequal these terms may be neglected giving the result:
2.2 Theory of Elasticity 29

h exx eyy ezz Dx =Dx Dy =Dy Dz =Dz For an isotropic body (physical properties are
independent of direction) and for an elastic body,
under small strain, strain varies linearly with the
The sign convention of (h) is negative for applied stress. This linear stress-strain relation-
compression and positive for expansion strains. ship is governed by a well-known mathematical
equation. It is the Hookes law.

2.2.7 Stress-Strain Relationship


2.2.8 Hookes Law for Isotropic
It is a common experience that a body under Media
stress undergoes deformation of a form and value
depending on the applied load and on the physical In its simple form, Hookes law states that the
properties of that body. Bodies of the type which, strain-stress relationship is linear. It is applicable
under stress, exhibit a proportional strain are to the behavior of stressed bodies when stresses
called elastic bodies. When the proportionality is are sufciently small. If several stresses are act-
linear, these are called perfectly elastic bodies. ing on a body, the net strain produced is the sum
Normally, bodies, under increasing stress, of the individual strains. This is one of the
exhibit linear stress-strain behavior up to a certain important outcomes of the linearity property of
stress-limit, beyond which the material may still the stress-strain relationship of isotropic media
be elastic but with no more linear relation- under small strains. A medium under stress
ship. Usually there is a point (the elastic limit) after condition, in which Hookes law holds, is called
which the deformation becomes irrecoverable and Hookean medium.
in this case the body behavior is described to be When the stressed bodies are isotropic (their
plastic. An increase of stress beyond the elastic physical properties do not change with direction)
limit produces large increase in strain, and it does the linear stress-strain relationship becomes rel-
so even with decreasing stress. With further atively simple linear function (Sheriff and Gel-
increase of an extensional stress (tensile loading) dart 1995, p. 37). The linear equation that
for example, a point is reached where the body can connects stress to strain of isotropic media is
no longer sustain the applied stress. At this point commonly found in the geophysical literature, as
(called the rupture point) the body breaks in Bullen (1965, p. 20), McQuillin et al. (1984,
up. Behavior of a ductile solid-body under an p. 11), Sheriff and Gildart (1995, p. 37).
increasing extensional stress is shown in Fig. 2.9.

Fig. 2.9 Elastic and linearity elastic rupture


plastic zones shown by a point
limit limit
solid ductile body under an
increasing tensile stress
stress

linear zone none-linear


Hookes
law

elastic zone plastic zone

strain
30 2 Seismic Waves

The linearity property governed by Hookes law 2.2.9 The Elastic Moduli
means that there is a proportionality-constant for the
linear stress-strain relation for any particular body The elastic modulus of a body is the propor-
under stress. Mathematical studies showed that, for tionality constant of the stress-strain linear rela-
an isotropic body, two elastic coefcients are suf- tionship. It expresses an important physical
cient (Richter 1958, p. 238). These are the Lames property which is the extent of resistance of that
coefcients (k & l), which are sufcient in char- body to the applied stresses. Moduli of important
acterizing the elastic properties of a medium. practical applications are Youngs Modulus, bulk
By use of Lames coefcients (k & l), modulus, and shear modulus. These are dened
Hookes law can be presented in the following in the following discussions.
compact form:
(i) Youngs Modulus and Poissons Ratio
Let a simple tensile stress (Tx) be applied to
Tij = ij+ 2 eij an isotropic bar placed along the x-axis. This will
cause the bar to experience a longitudinal
extension (ex) in the x-direction and, at the same
where the symbols (i & j) take the values x, y,
time, it experiences lateral contractions along y-
and z, and the term dij = 1 when (i = j), and
and z-directions. Being an isotropic body, the
dij = 0 when (i 6 j). T and e are the stress and
contractions in the y- and z-directions (ey, & ez)
strain respectively.
are equal. These changes (expressed by the
This compact form of the Hookes law can be
strains ex, ey, & ez) are governed by the elastic
presented in the following explicit equations:
coefcients of the stressed body. The coefcients
which govern the stress-strain relation, in the
Txx = + 2 exx presence of the tensile stress (Tx), are Youngs
Tyy = + 2 eyy modulus (Y) and Poissons ratio (r).
Tzz = + 2 ezz For a one-dimensional stress acting on a body
obeying Hookes law, Youngs modulus (Y) is
For pure shear strain (that is with no change in the proportionality constant in the linear relation
volume, for h = 0), the Law expresses the rela- that connects stress (Tx) with strain (ex). The
tions for purely shearing strain, that is: relationship is:

Txx = 2 exx Tx Yex


Tyy = 2 eyy In the case of a rectangular rod of length (L),
Tzz = 2 ezz cross-sectional area (DA) stretched by (DL) due
to force (F), Youngs modulus (Y) is given by
From this equation, it is evident that, the stress Fig. 2.10:
is consisting of the sum of two parts; the rst part
(k h), involving the elastic coefcient (k) multi- Y Tx =ex F=DA=DL=L
plied by the volume change (the dilatation, h)
Youngs modulus is measured by pressure
and the second part (2l eij) which is involving
units (as psi, dyne/cm2 or N/m2).
the second elastic coefcient (l) multiplied by
The Poissons ratio (r), on the other hand, is
the longitudinal strain (eij). The coefcients (k
dened as the ratio of transverse strain (ey or ez)
and l) are called Lames constants. These two
to longitudinal strain (ex). For an isotropic body,
constants (and other related constants) which are
this is given by:
representing proportionality constant between
stress and strain, are normally referred to as the
elastic coefcients, or elastic moduli.
2.2 Theory of Elasticity 31

The plus sign is entered to denote density


A F increase for increase in compression.
L L It is sometimes called (Incompressibility) and
its inverse (1/B) is called (Compressibility). Its SI
Fig. 2.10 An elastic rectangular rod under extension unit is the pressure measuring unit (the pascal).
force
(iii) The Shear Modulus
r ey =ex ez =ex The shear modulus, (l), which expresses the
relationship between shearing stress and shearing
The minus sign is used to indicate that (ey) strain, is dened as the ratio of the shear stress
and (ez) are contractions for elongation (ex). (Txy) and the shearing strain (exy) represented by
the resulting angular change. For tangential force
(ii) The Bulk Modulus (F) acting on the face of a rectangular block of
The Bulk modulus (B) is dened to be the ratio area (DA), the shear modulus (l) is dened as
of change in hydrostatic pressure (P), acting on follows (Fig. 2.12):
a solid body of volume (V), to the relative
decrease in its volume (V/V). For a cube of l Txy =exy F=DA=Dx=h
volume (V) under hydrostatic pressure-change
(P), the bulk modulus (B) is given by The strain (exy) in this case is tangent of the
(Fig. 2.11): angle of shear (), or the angle in radians for
small value of the angle (). That is,
B DP=DV=V V  DP=DV DP=h
l Txy =
The minus sign is entered to denote volume
The angle () is normally called angle of
decrease for increase in compression and (h) is
shear and the coefcient (l) is the shear modulus
the cubical dilatation:
or rigidity modulus as it is sometimes called.
The Bulk Modulus (B) is a measure for the
Measurement unit of the shear modulus is pres-
body resistance to uniform compression. An
sure units as in the case of Youngs modulus.
equivalent expression for the bulk modulus can be
It is to be noted here that (l) serves as mea-
given in terms of density change (q) instead of
sure for the resistance of an elastic solid body to
the volume change. Thus, the denition becomes:
shearing deformation (i.e. to shape changes) and
B qDP=Dq that is why it is called rigidity modulus. For this
reason, it is equal to zero for a fluid medium as it
has zero-resistance to shape-changes.
(iv) Lames Elastic Coefcients
P P
Lames coefcients (also called Lames parame-
ters) are two parameters (k & l) which are used

V
A F
P
P x
V-V
h
P

Fig. 2.11 An elastic cube under hydrostatic compression Fig. 2.12 An elastic rectangular block under shearing
forces force (F) acting on area (DA)
32 2 Seismic Waves

Table 2.3 The Modulus Relation-1 Relation-2


mathematical
interrelationships of elastic Youngs modulus (Y) Y = l(3 k + 2 l)/(k + l) Y = 9Bl/(3B + l)
moduli, for an elastic Bulk modulus (B) B = (3 k + 2 l)/3 B = Y/3(1 2 r)
isotropic body (Sheriff
Shear modulus (l) l = 3 (B k)/2 l = Y/2(1 + r)
1973, pp. 6970)
Lames modulus (k) k = (3B 2 l)/3 k = rY/(1 + r)(1 2r)
Poissons modulus (r) r = k/2(k + l) r = (3B 2l)/(6B + 2l)

in characterizing the elastic properties of an iso- elastic constants, as we mentioned above. In


tropic medium. These two coefcients give addition to the two Lames coefcients (k & l),
complete elastic characterization of homogenous the other moduli: Youngs modulus (Y), Bulk
and isotropic media. They serve as the propor- modulus (B), and Poissons ratio (r) can be used
tionality constants in the stress-strain linear in characterizing the elastic properties of an iso-
relationship which is mathematically expressed tropic body. Any of these moduli can be
by Hookes law. expressed in terms of two other moduli as it is
To understand the physical implication of the summarized in Table 2.3:
rst coefcient (k), let us assume a solid cube It is evident from this table that any one of the
being stretched by a tensile stress (Tzz) resulting in ve constants can be expressed in terms of any
a corresponding tensile strain (ezz). The lateral two of the remaining constants. This implies that
tensile stress (Txx) needed to prevent lateral con- any two of these three constants can be used to
traction is, according to (Sheriff 1969), given by dene the elastic properties of a homogeneous
(Txx = k ezz). This relation furnishes the formal and isotropic medium.
denition of the Lames coefcient (k). The second For most rocks, values of the moduli (Y, B, &
coefcient (l) is the shear (or rigidity) modulus. l) lie in the range (2  1010 12  1010) N/m2,
Both of Lames coefcients (k & l) are with (Y) being the largest and (l) the smallest of
functions of other elastic constants. For instance, these three (Sheriff and Geldart 1995, p. 38).
they are functions of Youngs modulus (Y) and Table of values of elastic moduli of rocks have
Poissons ratio (r). The relations are: been published by Birch (1966).

k rY=1  2r  1 r
2.3 Wave Motion Equation
l Y=21 r
If two neighboring points in a stressed medium
experience the same stress, no motion of one of
Other relations are presented in Table 2.3.
them will occur with respect to the other. How-
ever, relative motion will take place when there
is a stress difference. In other words, motion
2.2.10 The Elastic Moduli
occurs when there is a stress gradient. This
Interrelationships
reminds us of an analogous case we met in the
creation of strain (see Sect. 2.2). The two cases
For a homogeneous and isotropic medium under
may be expressed as follows: Displacement
stress, the stress-strain relationship is linear
gradient is required to create strain and stress
within the elastic (Hookean) state. The propor-
gradient is required to cause motion.
tionality constants are the elastic moduli or
2.3 Wave Motion Equation 33

2.3.1 One-Dimensional Scalar Wave This strain (exx = Dx/x) is produced by the
Equation corresponding stress gradient (Txx/x).
By making use of the fact that the net force
In this section, we shall deal with the wave acting on any face is given by the stress acting on
motion equation which expresses the motion of a that face times the face area, we get the resultant
disturbance in one dimension. The disturbance in force (Fx) in the x-direction due to the stress
this particular case is the scalar quantity, the change (Txx/x)  dx that occurred across the
cubical dilatation (h). distance (dx). This is computed as follows:
Let us consider an elementary parallelepiped
(of dimensions: dx, dy, Dz) located inside an Fx @Txx =@x  dx  dy  dz
elastic isotropic medium (Fig. 2.13).
At each face of this elementary body, when it By applying Newtons second law of motion
is under elastic stress, there exist three stress we can express (Fx) in terms of mass of the
components: one is normal and two are tangen- parallelepiped (dx  dy  dz times density q)
tial to the particular face. The three multiplied by acceleration (2Dx/t2) in the x-
stress-components (Txx, Tyx, and Tzx) are acting direction giving:
on the face perpendicular to the x-axis. Under 
dx  dy  dz  q @ 2 Dx =@t2 @Txx =@x  dx  dy  dz
elastic stress-strain conditions, each of these
components will have a gradient in the
x-direction (Txx/x, Txy/x, and Txz/x). For or:
a complete three dimensional state, additional 
similar gradients occur in the other two directions q @ 2 Dx =@t2 @Txx =@x
(y-direction and z-direction).
In order to simplify the mathematical deriva- This is the one-dimensional (dimension, x in
tion of the equation of seismic wave motion, let a this example) wave motion equation which
plane compressional seismic wave to be describes particle motion (displacement, Dx) in
advancing in the x-direction. In this case the terms of the applied stress (Txx). However, the
three stress-components are reduced to only one motion can be expressed in terms of displace-
component (Txx) creating the corresponding ment only. This is done by using the stress-strain
strain (exx). When a seismic plane wave propa- linear relationship expressed by Hookes law
gates in the x-direction, the two faces (perpen- equation for isotropic media (Txx = k h + 2l
dicular to the x-axis) of the parallelepiped will be exx). Substituting for (Txx), the previously-
unequally displaced, and hence, it is subjected to derived wave equation becomes:
an elastic strain (exx) which is, by denition, 
given by the displacement gradient (Dx/x). q @ 2 Dx =@t2 k@h=@x 2l@exx =@x

Fig. 2.13 A small


parallelepiped element of
volume under elastic stress
34 2 Seismic Waves

Since, by denition, (h = exx + eyy + ezz) and three-dimensional wave-motion equation is a


(eyy = ezz = 0) in this case (case of restricting the linear second order partial differential equation.
disturbance to be displacement in the x-direction, The general wave equation of a disturbance, q(x,
with no lateral contraction), we can readily write: y, z, t) moving in space with velocity (v), is
given by:

q @ 2 Dx =@t2 k 2l@exx =@x
@ 2 q=@t2 v2 @ 2 q=@x2 @ 2 q=@y2 @ 2 q=@z2 :
or,
  or (using Laplacian operator (2 q = 2q/
@ Dx =@t
2 2
k 2l=q @ Dx =@x
2 2
x2 + 2q/y2 + 2q/z2), this can be written as:

This is a partial differential equation of the @ 2 q=@t2 v2 r2 q


form (2y/t2 = v2 2y/x2), which has the
general solution, y(x, t) = f(x vt) + g(x + vt). In seismic waves there are two types of dis-
In analogy to this standard form of partial dif- turbance (q) which propagate through the Earth
ferential equation, we can write the solution of materials. These are the two forms of elastic
the one-dimensional wave equation as: strains which represent the scalar volume
changes (expressed by the cubic dilatation, h)
Dx x; t f x  vt gx vt; and the vector shape changes (expressed by the
where; v k 2l=q1=2 : shear strain, w).
(i) The Scalar quantity (h)
This solution represents one dimensional This is the scalar cubic dilatation (h). It repre-
wave equation, which is a disturbance (particle sents disturbance (volume-changes) that moves
displacement, Dx) moving with speed of (v) in in space with velocity (v), where v = [(k + 2l)/
the positive x-direction as expressed by the rst q]1/2. According to the wave motion equation
term, f(x vt). The second term, g(x + vt) rep- (see for example, Richter 1958, p. 658):
resents a wave moving in the negative
x-direction. q @ 2 h=@t2 k 2lr2 h

By denition, the cubic dilatation (h) is rela-


2.3.2 The Scalar and Vector 3D ted to displacement D (vector quantity of com-
Wave Equations ponents Dx, Dy, Dz) by the formula (h = Dx/
x + Dy/y + Dz/z). Using vector notation,
Solution of the one-dimensional wave equation, the scalar quantity (h) is, therefore, the diver-
q(x, t) expresses the variation of the disturbance gence of the vector (D). That is:
(q) along the travel distance (x) at any time (t). In
the three-dimensional case, we have the depen- h div D r  D
dant variable; q(x, y, z, t) which possesses, at any
time (t), a dened value at any point in the sur-
rounding space (x, y, z). (ii) The Vector quantity (w)
The standard wave equation describes the The second type of moving disturbance is the
strain-changes as function of space and time as it vector quantity (shear strain, w). It represents
propagates (with constant velocity) through a (shape-changes) which moves with velocity,
perfectly elastic medium. It can be shown (see v = [l/q]1/2. The three components of the vector
for example Richter 1958, pp. 657658; Sheriff (w) are (wx, wy, wz), dened by (Richter 1958,
and Geldart 1995, pp. 3940) that the standard p. 658):
2.3 Wave Motion Equation 35

 represented by a plane normal to the x-axis.


wx @Dz =@y  @Dy =@z
Further, when the moving disturbance f(x vt) is
wy @Dx =@z  @Dz =@x
 in the form of sinusoidal function the moving
wz @Dy =@x  @Dx =@y disturbance is referred to as (plane harmonic
wave). Such a function is:
Each of these three components moves with
velocity (v) in accordance to the following f x; t a cos kx  vt;
wave-motion equations.
where (a & k) are constants.
q@ 2 wx =@t2 lr2 wx According to Fourier Theorem, a moving
pulse of an arbitrary shape can be transformed
q@ 2 wy =@t2 lr2 wy
into its harmonic components by superposing
q@ 2 wz =@t2 lr2 wz many sinusoidal functions, each of which is of
the form:
It is clear from the denitions of the compo-
nents of the vector quantity (w) that each of the f x; t a cos kx  vt;
components (wx, wy, wz) is the curl of the cor-
responding displacement-components (Dx, Dy, or,
Dz) as expressed above. That is:
f x; t a cos 2px=k  t=s
wx curlx D r  D
This equation is showing that f(x, t) is a pe-
Similarly, for the other two components riodic function of (x & t) which is oscillating
(wy, wz). with wavelength (k) and period (s) in respect to
distance and time respectively, where k (=2p/k)
is called the wave number. The factor (a) repre-
2.3.3 Plane Waves sents the amplitude of the particular harmonic
component. The two forms of these two equa-
A plane wave is dened as that wave for which tions are equivalent since v = k/s.
the moving disturbance is constant at all points of Periodicity of the harmonic plane wave is
any plane perpendicular to the propagation expressed by two parameters. These are the
direction (Fig. 2.14). spatial frequency in cycle per meter (fx) and the
For the seismic plane waves, the elastic dis- temporal frequency in cycle per second (ft).
turbances, h(x, t), or w(x, t) are functions of the These are related to the wavelength (k) and to the
travelled distance (x) only. In either of these two wave period (s) by the relations (fx = 1/k) and
types of disturbances, therefore, the wave front is (ft = 1/ s) respectively.

Fig. 2.14 Wave fronts of z


a plane wave advancing in y
x-direction

x
plane wave fronts
36 2 Seismic Waves

2.3.4 The P- and S-Waves 330 m/s in air, 1450 m/s in water, and (2000
6000) m/s in rocks.
As it is presented above, there are two types of A solid medium having its Poissons ratio
disturbance that can move in accordance with the equal to (1/4) is called Poissons solid (Sheriff
standard wave motion equation. These are the 2002, p. 266).
scalar cubic dilatation (h) and the vector shear
strain (w).
From the wave equation it can be shown that 2.4 Classification of Common
the disturbance (h) moves faster than the other Elastic Waves
disturbance (w). Thus, when the two distur-
bances are generated by a certain source, the From analyses of stress and strain, we have seen
(h-wave) arrives earlier than the (w-wave). For that strain is, in general, made up of two types of
this reason the two waves are called Primary elastic disturbance; the cubic dilatation and the
(P-wave) and Secondary (S-wave) respectively. shear strain. Solution of the equation of motion
It is to be noted that the ratio of the P-wave showed that each of these types of deformation
velocity (vp = [(k + 2l)/q]1/2) to the S-wave travels through the medium with its own veloc-
velocity (vs = [l/q]1/2) is equal to [(k + 2l)/ ity. The rst type of disturbance represents the
l]1/2. Using the relationship connecting (k) and moving volume strain and the second type
(l) in which k/l = 2r/(1 2 r), we can write: involves the shape strain. The rst type is
called Longitudinal, Compressional, or Primary
vp =vs 2  2r=1  2r1=2 wave (or just P-wave) which travels faster than
the second type which is called Transverse,
This formula clearly shows that the ratio of Shear, or Secondary wave (or just S-wave).
the P-wave velocity (vp) to the S-wave velocity These two types of waves (P- and S-waves)
(vs) is function of Poissons ratio (r) only. belong to a class of waves (called body waves)
According to (Dobrin 1960, p. 18), Poissons because they can propagate through the interior
ratio (r) generally ranges from 0.05 to 0.40, of the earth body. This group of waves is called
averaging about 0.25 for hard rocks. With this so to differentiate them from another class of
value (r = 1/4), the velocity ratio (vp/vs) waves which move on and near the free surface
becomes 31/2 (=1.732). This means that P-wave of the medium, called (surface waves) which
moves with velocity which is about 1.7 times as include Rayleigh- and Love-waves. Classica-
fast as the S-wave moving in the same medium. tion of the common elastic (seismic) waves is
It is useful to note that P-wave velocity is shown in Fig. 2.15.

Fig. 2.15 Classication of


the common seismic waves Seismic
Waves

Body Waves Surface Waves

Longitudinal Transverse Rayleigh Love


(P-Waves) (S-Waves) Waves Waves
2.4 Classification of Common Elastic Waves 37

2.4.1 Body Waves P-wave is the fastest wave for a given medium
and, therefore, its arrival at a certain observation
Body waves are waves that can travel through an point is the earliest among the seismic
elastic materialistic medium in any direction. As wave-types. This is a common observation of
they move, the waves may experience changes in seismologists working on analysis of earthquake
their energy level and in their travel-path geom- seismograms. Propagation velocity (vp) of
etry subject to the physical properties of the P-wave depends on the medium density (q) and
medium. There are two sub-types of these waves; elastic properties (k & l) and it is given by the
the longitudinal and the transverse waves expression vp = [(k + 2l)/q]1/2.
(Fig. 2.16).
(ii) Transverse Waves
(i) Longitudinal Waves The travelling disturbance in this case is the shear
This type of waves is also known as compressional, strain or shape deformation. The medium
Primary, or just P-wave. The travelling disturbance which is traversed by this type of waves experi-
in this case is volume deformation expressed by ences no volume changes. A consequence of the
the cubical dilatation (h) as dened above. shear strain (rotation of part of the medium) is the
The particles of the medium, traversed by a transverse displacement of the path particles
plane P-wave, vibrate about their neutral posi- relative to the propagation direction. They are
tions in the direction of the wave propagation. also called (shear waves) or (Secondary, or just
The travel path consists of a sequence of alter- S-waves).
nating zones of compressions and rarefactions A horizontally moving S-wave, which is so
(Fig. 2.16a). This is the type of waves which is polarized that the particle motion is conned to
commonly employed in seismic reflection and vertical plane, is known as SV-wave (Fig. 2.16
refraction exploration work. b). When the polarization plane is horizontal, it is

Fig. 2.16 Particle (a)


displacement-mode of a
medium traversed by plane
body-waves, k is
wavelength. a P-wave, propagation
b SV-wave, c SH-wave direction

(b)
SH-
propagation
direction

(c)

propagation
direction
38 2 Seismic Waves

called SH-wave (Fig. 2.16c). The velocity of The main sub-types of surface waves are
S-waves, vs is given by vs = [l/q]1/2. In Rayleigh waves and Love waves (Fig. 2.17).
liquid-media, where (l = 0), S-waves do not
(i) Rayleigh Waves
propagate.
Rayleigh waves, which were discovered by an
English scientist, Lord Rayleigh in 1885, are
usually developing at the free surface of a
2.4.2 Surface Waves
semi-innite solid medium. Its wave amplitude
decays rapidly with increasing depth. The trav-
As it is implied by its name, surface waves are
elling disturbance in this case is a sort of com-
waves that move on the free surface of the earth.
bination of particle-motions of both P- and
The main features common among all surface
SV-waves. The particle motion, which has a
waves, observed on earthquake seismograms, are
retrograde elliptical orbit, takes place in a vertical
their relatively large amplitudes (high energy
plane parallel to propagation direction
content) and low frequencies when compared
(Fig. 2.17a). The minor axis of the elliptical orbit
with the body waves. In addition to that, they
is parallel to wave motion direction and it is
move with velocity which is generally slower
equal to two-thirds of its major axis. Rayleigh
than body waves moving in the same medium. It
waves travel on the surface of a solid medium
is a common observation that the dispersion
with velocity of 0.92 of the velocity of S-waves
phenomena are more prominent in surface waves
moving in that medium (Bullen 1965, p. 90). In a
due to dependence of the velocity on the fre-
sense, Rayleigh waves are similar to the familiar
quency of individual harmonic component.
water waves, with a fundamental difference, and

(a)

propagation
direction

(b)

surface layer

propagation
direction

Fig. 2.17 Particle displacement-mode of a medium traversed by plane surface-waves, (k) is wavelength. a Rayleigh
Wave, b Love Wave
2.4 Classification of Common Elastic Waves 39

that is the particle motion in case of Rayleigh Since they possess no vertical component,
waves describe an elliptical path whereas the Love waves are not detected by the geophone or
particle-motion path in case of water waves are by any such-like vertical-component sensing
circular in shape. instrument.
In the case where the semi-innite medium is
overlain by a low-velocity surface layer, Rayleigh
waves exhibit a phenomenon known as (disper- 2.4.3 Seismic Noise
sion). Harmonic components of longer periods
(lower frequencies) travel faster. Consequently, Broadly speaking, the term (noise) used in seis-
the Rayleigh wave seismograms would, in gen- mology, is applied to all types of disturbance
eral, show decrease in period along the wave-train. which may interfere with (and impose masking
Components of too-long wavelengths (too long effects to) the seismic signal of interest. In this
compared with the thickness of the surface layer) way, the concept of seismic noise bears a relative
penetrate deeper and travel with velocity of about implication. Thus, when the interest is focused
0.9 times the S-wave velocity in the subsurface on reflected body waves, surface waves and other
material. The short wavelengths travel mainly in non-reflection waves (as direct and refraction
the surface layer with velocity of about 0.9 times arrivals) are considered to be the unwanted
the S-wave velocity in the surface layer. troublesome noise. If the interest is in the
Surface waves, normally seen on shot records, refraction arrivals, reflection arrivals become
obtained in seismic reflection surveys, are com- the unwanted noise. In the strict sense, however,
monly called (ground roll) and these are identied the ambient seismic disturbances (usually of
to be of Rayleigh-wave type. Sometimes, these are random energy distribution which form the
called pseudo-Rayleigh waves (Sheriff 2002). background of a distinct travelling signal) are
Ground-roll waves are considered to be unwel- considered to be the seismic noise.
comed noise and efforts are usually made to get rid Seismic noise has destructive effects on the
of them or at least minimize their masking effect seismic signals of interest. A signal recorded
caused to the seismic reflection signal. amid a background of noise is distorted and
weakened because of the interfering noise. Signal
(ii) Love Waves
resolution is badly affected with noise develop-
This is the second sub-type of surface waves which
ment. A measure for the signal resolution, called
was discovered, in 1911, by another English geo-
the signal-to-noise ratio (S/N ratio) is usually
physicist named A.E.H. Love (18631940). It
applied. It is dened to be the ratio between
develops only in cases where a solid elastic
signal amplitude detectable amid a background
semi-innite medium is overlain by a horizontal
seismic noise.
low-velocity layer. Like SH-wave vibration mode,
In exploration seismology seismic noise is
the particle movement is transverse and is conned
divided into two main types; coherent and inco-
to the horizontal plane (Fig. 2.17b). Love waves
herent noise (Fig. 2.18).
travel by multiple reflections between the top and
bottom boundary-planes of the surface layer. The (i) Coherent Noise
propagation velocity approaches S-wave velocity Coherent noise is a seismic event characterized by a
in the subsurface medium for very long wave- distinct apparent velocity and well-dened onset.
lengths and to that of the surface layer for short In reflection seismology, coherent noise which
wavelengths (Dobrin 1960, p. 23). Love waves appear on shot records, are source-generated seis-
always exhibit dispersion. As in the case of Ray- mic events. They are made up mainly of surface
leigh waves, Love waves propagation-velocity waves (ground roll) and air-waves which are of
increases with the period of the harmonic compo- fairly narrow bandwidth with low frequency range.
nent. Again, the vibration amplitude decays expo- Frequency content of this type of noise is typically
nentially with depth in the lower medium. below 20 Hz (Fig. 2.18).
40 2 Seismic Waves

Fig. 2.18 Sketch showing (a) ground roll


amplitude spectra of
common coherent noise,
ground roll (a), and
(b) seismic reflection signal

amplitude
incoherent noise, random
noise (c), in relation to that
of the reflection signal (b)

(c) incoherent noise

0 20 40 60 80 100
frequency, Hz

(ii) Incoherent Noise at, in seismic data acquisition. Several ways and
Unlike coherent noise, the incoherent noise means are followed in the eld-acquisition stage
consists of seismic events with unpredictable or in the following data-processing stage to get
amplitude and onset. This type of noise, which is enhanced S/N ratio. Suitable measures are
basically of random nature, forms the applied to the parameters of the seismic source
seismic-energy background of any seismic and detectors as well as those measures applied
shot-record. In earthquake seismology it is in processing work, in order to attenuate these
commonly known as (microseisms), and in noises and enhance the S/N ratio.
prospecting seismology it is called (incoherent
background noise), or (ambient noise) as it is
sometimes referred to. In addition to the ran- 2.5 Propagation of Seismic Waves
domness nature, the incoherent noise is charac-
terized by a broad amplitude spectrum that Seismic waves are generated from a sudden
covers a wide range of frequencies compared change in the internal strain occurring inside an
with the nearly limited bandwidth of reflection elastic medium. The generating source may be
signals or coherent noises (Fig. 2.18). In the natural as in the case of earthquakes or articial,
geophysical literature we sometimes meet terms like exploding a charge of dynamite, as nor-
like (white noise) indicating wide-band noise, mally done in seismic exploration. All parame-
and (red noise) for low-frequency random noise. ters of an advancing seismic wave (waveform,
Intensive research work has been undertaken, speed, and travel-path geometry) may change
directed towards a greater understanding of the during the wave propagation. Form and mag-
source and characteristics of the incoherent nitude of these changes depend on the physical
noise. It is now generally accepted that it is properties of the host medium. Whether the
generated as a result of external energy sources source is natural or articial, a seismic eld is
like wind movements, sea-waves collisions with created when a sudden pressure pulse is initi-
sea coasts, in addition to other various natural ated. The generated seismic energy moves away
and articial man-made activities. from the source zone in a form of a wave
Because of seismic noise which are unavoid- motion propagation. Under these conditions
able seismic events which get recorded alongside (seismic energy source within an elastic med-
the objective signal, the signal-to-noise-ratio ium), the seismic wave spreads out from the
(S/N) becomes an important parameter in signal source zone in every possible direction. A travel
detection studies. The S/N ratio is used as mea- ray-path, in a particular medium, is dened once
sure for the signal quality-level. Signal clarity the locations of both of the source-point and
(S/N enhancement) is a central objective, aimed detector-point are dened.
2.5 Propagation of Seismic Waves 41

spreads out into the three dimensional space of


homogenous elastic medium the host medium. If the medium is homogeneous,
the seismic energy would advance in every
source seismic pulse possible direction with constant velocity. This
means that after any given travel-time the energy
ray-path would have reached points of equal distances
detector from the source. These points fall on a spherical
surface which is marking the (wave-front). For a
Fig. 2.19 Elements of the seismic eld, shown for an harmonic seismic wave, the wave front is dened
idealized homogenous and elastic medium to be the locus of points having the same phase
of particle vibration.
2.5.1 Elements of the Seismic Field At any point in the wave-eld, the line which
is perpendicular to the wave front at a certain
In case of an idealized simple medium (one instant represents a (ray). The ray is an imagi-
which is homogeneous, isotropic, and perfectly nary line normal to the wave front at a certain
elastic medium) the wave-motion propagation is point which indicates the motion-direction of the
expected to be along straight ray-paths, with advancing wave at that point. Near the source
constant velocity. A seismic eld is created when point, the wave fronts of seismic waves travelling
a mechanical energy within an elastic medium through a homogeneous medium are of spherical
generates a seismic pulse that propagates through shapes and thus the rays are straight lines radi-
that medium. The fundamental elements of a ating in all directions from the source point. At
seismic eld are, thus, a source of mechanical very large distances, the wave fonts are approx-
energy, elastic medium, and detector. imating to planes and the rays, in this case,
Except for the geometrical spreading effect, become parallel straight lines perpendicular to
the wave moves through such an idealized the plane wave-fronts.
medium, with no changes taking place on The familiar example is the wave which
ray-path direction or on the waveform of the develops on the surface of water when a small
travelling seismic pulse (Fig. 2.19). solid object (a pebble, say) is dropped vertically
into it. The crests and troughs of the generated
wave spread out from the source-point in the
2.5.2 Concepts of Wave-Fronts form of concentric circles. In fact these circles
and Rays are depicting the surface expression of the
spherical wave-fronts which are advancing
From a mechanical energy-source, (such as a through the three-dimensional space of the water
mechanical pressure pulse), a seismic wave medium. By denition, the ray at any point on

Fig. 2.20 Concepts of the


wave fronts and rays as
seen when a water wave is
created from dropping a
pebble into a still pond

ray

wave-front
42 2 Seismic Waves

the wave front is a line drawn normal to the 2.5.4 The Concept of the Interface
spherical wave-front (circles on the surface
plane) at that point. Concepts of the wave-front The Interface is that boundary-surface separating
and rays are shown in (Fig. 2.20) for a case of two different media. As far as the changes (chan-
dropping a pebble into a still pond. ges in spectral structure and propagation direction)
of seismic waves are concerned, two media are
considered to be different if both of the wave
2.5.3 Huygens Principle propagation velocity and the medium bulk density
are different. Since velocity is function of elastic
Huygens Principle states that each point on a coefcients, it can be said that density and elastic
wave-front acts as a source of a new wave which, properties are the factors which control the specic
in a homogeneous medium, generates a sec- characters of the media. The parameter which
ondary spherical wave-front, the envelope of expresses the combined effect of velocity and
which denes the position of a wave generated at density is called (acoustic impedance) which is
some later time. dened to be the product of velocity by the density.
Huygens model of wave propagation requires To clarify the concept of the interface and the
that the secondary wave-fronts are active only at roll of the acoustic impedance waves hitting an
the points where the envelope touches their sur- interface let us consider a two-layer model which
faces. The wave energy is spreading out from the consists of two adjacent media (M1 & M2) of
primary source-points in all directions, but their velocities and densities (v1 & q1) for medium (M1)
mutual interactions make the resultant distur- & (v2 & q2) for medium (M2). The acoustic
bance zero everywhere except at the points impedances (z1 & z2) for the two layers are
where they touch the common envelope. (z1 = q1v1) and (z2 = q2v2) as shown in Fig. 2.22.
Applying the principle on plane-wave propaga- In analogy to the role of electrical impedance
tion in a homogeneous, and in an inhomoge- in the flow of electrical current, the acoustic
neous medium, is shown in Fig. 2.21. impedance expresses the extent of resistance the

Fig. 2.21 Plane-wave (a) (b) (c)


propagation according to two separate media
Huygens Principle. homogeneous medium one medium of
no change in velocity of different
a Through a homogeneous uniform velocity
velocities
medium where velocity is change
constant. b Inhomogeneous
medium of velocity which
is uniformly changing
across the propagation medium-1
direction. c Two media of V1
different velocities pla
ne medium-2
V2

direction of two media of


medium of constant
increase of velocity, V velocities, V1 < V2
velocity, V
2.5 Propagation of Seismic Waves 43

Z1 = 1 V1 M1
Interface plane

Z2 = 2 V2 M2

Fig. 2.22 Concept of the Interface and denition of the Acoustic Impedance, z (=qv)

seismic energy meets when traversing a medium. place on ray-path direction or on the waveform of
The higher the acoustic impedance, the lower the the travelling seismic pulse. In nature, however,
particle vibration-velocity will be, and vice versa. the medium is far from this idealized form. In the
Acoustic impedance is measured by (kg s1 m2) solid crust of the Earth, it is commonly made up of
or by the equivalent (Ns m3) units. rock layers of varying physical properties and
At an interface, an incident seismic wave varying geometrical forms and sizes.
(normally a P-wave in seismic exploration work) In such inhomogeneous environments a
would, under certain geometrical conditions, give moving seismic wave would suffer from a num-
rise to wave conversion in addition to reflection, ber of changes whenever it meets an interface
refraction, and diffraction. These cases shall be across which there is change in the properties of
dealt with in some details in the following the medium. In particular, changes in energy
discussions. content, waveform (spectral structure), propaga-
tion velocity, direction of motion, and new wave
generation. These changes, are generated at the
2.5.5 Changes of Propagation interface planes dening the layer bounding
Direction at Interfaces surfaces (Fig. 2.23).
The common changes in ray-path direction,
In an idealized homogenous and elastic medium, a which are of signicance to exploration seis-
seismic wave propagates with no changes taking mology, are: reflection, refracted transmission

Fig. 2.23 a Innite,


elastic homogeneous (a) (b)
medium showing straight
ideal medium realistic medium
ray-path. b Inhomogeneous
(layered) medium showing (homogeneous) (inhomogeneous)
changes in ray-path
direction source source detector

Layer-1

Layer-2

Layer-3

Layer-4
44 2 Seismic Waves

Fig. 2.24 Ray-path surface surface surface


geometry of the three most
common wave propagation
ray-paths: reflection,
refraction, and diffraction

V1 V1 V1
V2 V2 V2

reflection refraction diffraction

(refraction), and diffraction. These shapes of the of different density and elastic properties (dif-
moving wave ray-path occur at the boundaries of ferent acoustic impedances), four new wave
media having different seismic propagation phases are generated; reflected and refracted P-
velocities (Fig. 2.24). and SV-waves. If, however the incident is
SH-wave, the generated waves are only reflected
and refracted SH-wave. The SV-waves, gener-
2.5.6 Wave Conversion at Interfaces ated from an incident P-wave, (or P-waves gen-
erated from an incident SV-wave) are called
When a seismic wave impinges on an interface (converted waves) (Fig. 2.25).
separating two media, which differ in acoustic An incident seismic wave onto an interface
impedances, the incident seismic energy is partly will be partly reflected and partly transmitted
reflected and partly transmitted with certain across the interface. In general, the interface will
waveform changes. When the ray-path of an bring about wave conversion, reflection, trans-
incident seismic wave is oblique, that is inclined mission, and diffraction. It should be noted here
with respect to an interface, new waves are that refraction is a special case of transmission.
generated. If, for example, the incident wave is Refraction (ray-path bending) occurs only in the
P-wave (or SV-wave) separating two solid media case of inclined incidence.

P SV SV SH
P SH
SV
P

M 1 , V1 , 1 M 1 , V1 , 1 M 1 , V1 , 1
M 2 , V2 , 2 M 2 , V2 , 2 M 2 , V2 , 2
P P

SV SV SH

V2 > V 1

Fig. 2.25 Wave conversion at an interface for three types of incident waves (P, SV, and SH waves). The symbol (h)
denotes angle of incidence
2.5 Propagation of Seismic Waves 45

2.5.7 Energy Partitioning incident waves (P, SV, SH) and impedances
and Zoeppritz elements (velocity and density) for each of these
Equations wave-types. A typical set of Zoeppritz curves for
the case of an obliquely incident P-wave is
The mathematical expressions describing the shown in Fig. 2.26.
energy partitioning of an obliquely incident wave Referring to Fig. 2.26, it can be seen that, for
among the created converted waves, were normal incidence (angle of incidence = 0), no
derived rst by Knot (1899) and later on by S-wave is generated and thus all the energy is
Zoeppritz (1907), but not published until (1919). shared by the reflected and transmitted (re-
Using an approach (based on displacement fracted) P-wave. At a small angle of incidence,
computations), Zoeppritz has derived the equa- the converted S-waves are of small energy level.
tions (commonly known as Zoeppritz equations) As this angle increases the generated S-waves
which express the relative energy partitioning as grow stronger at the expense of reflected and
function of angle of incidence and acoustic refracted P-waves. At the critical angle of the
impedances of the media separated by the incident P-wave, the transmitted P-wave energy
involved interface. falls to zero, and at the same time, both of
In the geophysical literature, these equations reflected P-wave and reflected S-wave grow
are presented in the form of curves for certain large. Buildup of energy of reflected P-wave, as
two-layer models with dened density and elas- the critical angle is approached, is referred to as
ticity properties (see for example, Grant and (wide-angle reflection). This phenomenon (in-
West 1965, p. 54, Telford et al. 1990, p. 157, crease of reflection coefcient near the critical
Sheriff 2002, p. 401). A complete coverage of angle) is sometimes made use of in seismic
various types of incident waves, with different reflection exploration (Sheriff 1973, p. 241).
types of media, is found in (Ewing et al. 1957). Further, as the angle of incidence approaches
Because of the numerous possible grazing incidence (angle of incidence = 90),
parameter-values required to dene the behavior energy of the reflected P-wave increases and, at
of energy-distribution as function of incidence grazing incidence (where there is no vertical
angle, many curves are required for the various component for the incident P-wave), the S-waves
cases. These cases represent selected types of disappear and no transmission process occurs

1.0

relative
energy transmitted P-wave
reflected P-wave
reflected SV-wave
transmitted SV-wave

0.0 0 critical angle of incidence 90


angle

Fig. 2.26 Typical Zoeppritz curves of energy partition- P-wave at an interface separating two media of specied
ing as function of angle of incidence. The curves are for properties (sketched, based on Dobrin and Savit 1988,
the converted waves created by an oblique incident p. 43)
46 2 Seismic Waves

and consequently all the incident energy is con- case of reflections from a given horizontal
ned to the reflected P-wave. reflector. There are, however, situations where
the angle of incidence does not vary with the
offset. Thus, in a multi-reflector case, the angle
2.5.8 Amplitude Variation of incidence (which is equal to angle of
with Angle of Incidence reflection for the same wave-type), varies with
reflector depth for a xed offset. Also, in certain
For non-normal incidence, an incident P-wave cases, it is possible to get different offsets for a
leads to wave conversion in which both reflected xed value of reflection angle. These two cases
and transmitted P- and S-waves are sobtained. which occur in multi-reflector situation are
The obliquely-incident wave energy is dis- shown in Fig. 2.27.
tributed among all the converted waves in such a As expressed by Zoeppritz equations, the
way depending on the properties of the involved reflection coefcient shows variation with
media on both sides of the interface. According increasing angle of incidence (or with increasing
to Zoeppritz equations, the reflection coefcient offset). Depending on the distribution of the
is function of rock properties (density and elastic acoustic impedance on both sides of the inter-
properties) in addition to the angle of incidence. face, the reflection coefcient can vary from
For a given reflector, the amplitude variation large-negative to large-positive values. This
with angle of incidence (AVA) of a reflected behavior (variation of reflection coefcient with
seismic wave is found to be dependent on Pois- angle of incidence) can therefore be used as an
sons ratio as well as on impedance contrast indicator to predict lithological changes or type
across the reflection interface. In this way, the of fluid deposits.
parameter (AVA) possesses the same informa-
tion contained in a combined P- and S-waves
data. 2.6 Effect of the Medium on Wave
It is important to note that Zoeppritz equa- Energy
tions give direct relation of amplitude variation
with angle of incidence (AVA) and not ampli- Due to the earth ltering effect and other causes,
tude variation with offset (AVO). However, the wavelet generated by the source energy, is
offset is proportional to angle of incidence, in changed from its initial high-energy, impulsive

(a) surface
(b) surface

i1
i1
reflector-1 reflector-1

i2 i2
reflector-2 reflector-2

i3 i3
reflector-3 reflector-3
i1 > i2 > i3 i1 = i2 = i3

Fig. 2.27 Variation of reflection angle with reflector with increase of reflector depth for xed offset, and
depth for a xed offset and variation of receiver offset for b angle of incidence is constant for varying offset
a xed reflection angle. a Angle of incidence (i) decreases
2.6 Effect of the Medium on Wave Energy 47

Table 2.4 Factors 1. In the source zone 2. In the path zone 3. In the detector zone
contributing to energy
changes of a travelling Energy-Source parameters Reflection coefcient Detector response
seismic signal Source coupling Geometrical spreading Detector coupling
Near-source geology Inelastic attenuation Near-detector geology
Source-generated noise Wave conversion Surface noises
Noises and interferences Noises and interferences Noises and interferences

Fig. 2.28 Reflection source signal reflected signal


signal playground. The
source impulse getting
weaker and broader as it
progresses along its
reflection travel-path. surface
Factors affecting the signal
prevailing in the three
zones: source-, path-, and 2.path
1. source zone 3. detector
detector-zone
Source parameters Reflection Detector response
Source coupling coefficient Detector coupling
Near-source geology Geometrical Near-detector geology
Source-generated noise spreading Noises & interferences
attenuation

reflector

form into a lower energy and stretched-form 2.6.1 Geometrical Spreading


wavelet when observed at the end of its
travel-path. Taking the case of reflection of a In case of a homogeneous medium, seismic energy
seismic signal, the complete signal play-ground generated at the source, spreads out as spherical
and the main factors, contributing to the signal wave fronts concentric at the source point. Due to
energy changes, are summarized in Table 2.4 expansion of the advancing wave-front, wave
and Fig. 2.28. energy is distributed over increasing wave-front
We have already discussed the energy chan- surfaces. Mechanism of reduction of the wave
ges (expressed by the reflection and transmission energy level with travel-distance can be presented
coefcients) due to incidence of a seismic wave with the aid of (Fig. 2.29).
onto an interface. Two other important types of Referring to Fig. 2.29, let a source energy
energy changes due to the medium through (E) be generated at the source point, then, after
which the wave is propagating are to be dis- travelling distances (r1) and (r2) the corre-
cussed. These are the geometrical spreading and sponding energy-density of the spherical wave
the inelastic attenuation effects. fronts will be (e1) and (e2) respectively. The same
48 2 Seismic Waves

2.6.2 Inelastic Attenuation


e1
e2
r1 r2 Due to friction between the vibrating particles of
a medium traversed by a propagating seismic
E wave, some of the vibration energy is lost as a
result of being converted into heat. The amount of
loss increases with the increase of distance
(r) from source point. Experimentally, this is
Fig. 2.29 Wave-fronts generated by a point source are
spreading out as spherical wave-fronts in a homogenous
found to take an exponential function of the form:
medium ar ao ear ;

energy quantity (E) is distributed over the or,


wave-fronts of radii (r1 and r2), hence: at ao ea:vt:t

E 4pr1 2  e1 4pr2 2  e2 where, a(r) is the amplitude at distance (r), (a0)


is the initial amplitude, (a) is the attenuation
giving, coefcient (expressing amplitude reduction due
to absorption), (v) propagation velocity,
e1 =e2 r2 2 =r1 2 (t) travel-time, and (e) is base of natural loga-
rithm (=2.71828).
or, (since energy is function of the square of Earthquake seismologists often express the
amplitude, a), attenuation function in a different form (see for
example Bth 1974, p. 275),
a1 =a2 r2 =r1
at ext=2Q ;
This result has shown that amplitude attenu-
ation due to spreading of the wave-fronts (called Comparing this form with the form involving
geometrical spreading) is proportional to the the absorption coefcient (a), the a-Q relation-
travelled distance. In isotropic media, the energy ship is obtained. That is:
spreads out in the form of advancing spherical
surfaces. For this reason the phenomenon is a pf=Qv
sometimes called (spherical divergence).
The amplitude value is related to the travelled This means that the absorption coefcient (a)
distance (r) according to inverse relation. Thus, is linearly dependant on frequency, implying that
(a) is proportional to (1/r), or to (1/v(t)  t), higher frequencies are attenuated faster with
where v(t) is the velocity expressed as function increasing distance (or with time). This is sup-
of travel-time (t). For a medium made up of ported by experimental evidences, which are
parallel layers, it was shown by Newman (1973) proving that the earth is acting as a high-cut (or
that geometrical spreading depends on (1/v2(t)  low-pass) lter to travelling seismic waves. That
t) and not on (1/v(t)  t), that was derived for is why frequencies decrease with depth, as it is
homogeneous media. commonly observed on raw reflection-records.
It is important to be aware that geometrical The parameter (Q), called the quality factor,
spreading is independent of frequency. expresses the absorption-capability of the
2.6 Effect of the Medium on Wave Energy 49

medium. It is dimensionless quantity and inde- parameter is dened to be the natural logarithm
pendent of frequency. The quality factor is inver- of the ratio of two neighboring amplitudes of a
sely proportional to the attenuation factor (a). The gradually fading wave-train. This is customarily
term 1/Q is called the specic dissipation. measured by the ratio of two amplitudes sepa-
rated by one wavelength (Fig. 2.30).
By denition, the logarithmic decrement (d) is
2.6.3 Seismic Wave Energy given by:
Measurement
and the DB Unit d lna1 =a2 ln ear =ear k ln eak

We are all familiar with the units with which hence,


physical quantities are measured. Common
examples are: gram for measuring mass, meter for d ak
lengths, and seconds for time. Ratios, on the other
From this result the mathematical relations
hand, have no units as such. The decibel unit (or
connecting (a, Q, and d) can be readily obtained.
db), which is one-tenth of the bell unit, is intro-
Thus,
duced for measuring values of ratios in just the
same way as in measuring masses, lengths, and
a d=k pf =Qv;
other physical quantities. The db unit seems to
have been developed in connection with mea- Due to these natural attenuation effects, the
suring energy- or power-ratios of sound-wave reflection arrivals from deep reflectors are much
intensity expressed by its wave energy or by its weaker than those coming from shallow reflec-
wave amplitude. Likewise, the db-unit is usually tors. If a raw seismic trace is displayed, the
used in measuring seismic wave energy. reflection-events from deep reflectors are so
The decibel is dened to be the unit of mea- weak that they can be barely noticeable. How-
suring a power (energy) ratio (E), expressed in ever, when the attenuation due to spherical
logarithmic domain to the base 10, hence, divergence (geometrical spreading) and due to
absorption, is compensated, all events (shallow
Edb 10 log10 E20 log10 A
and deep events) can all be clearly seen
(Fig. 2.31).
where the power quantity (E) is related to the
square of amplitude (A).
From this denition, it is apparent that the
2.6.5 Wave Dispersion
ratio expressed in decibels is positive when
E > 1, and negative for E < 1, and it is zero
Wave dispersion is a phenomenon that occurs to
when E = 1. Another useful note is that ratios (in
the propagating wave for which velocity is
the db-domain) are added or subtracted corre-
sponding to multiplication or division of the
original ratios. For example the ratio of (2/1) in
db units, is 20log(2) which is equal to (6 db), and
that of the ratio (1/2), it is (6 db). a
a1
a2
x
2.6.4 The Logarithmic Decrement 0

There is an attenuation parameter, called the = ln (a1 /a2 ) = ln e- r/ e- (r+ )


, = ln e
logarithmic decrement (d), closely associated
with the inelastic attenuation coefcient (a). This Fig. 2.30 Denition of the logarithmic decrement (d)
50 2 Seismic Waves

Fig. 2.31 Seismic trace


before application
(unscaled) and after
application (scaled) of the
spherical divergence and
inelastic attenuation
corrections

function of the frequency component of the travelling wave, is moving. The wave-train or the
travelling wave. Dependence of velocity on fre- energy package (expressed by the envelope of
quency means that each frequency component of the wave train) is travelling with different
a seismic signal moves with its own velocity. velocity called (group velocity, U) as shown in
Thus a wave, composed of several Fig. 2.32.
frequency-components will experience The group velocity (U) is mathematically
component-separation, and hence, related to the phase velocity (V) and wavelength
change-of-form that occurs during travel. Dis- (k) of the frequency component, by:
tortion of the wave-form due to dependence of
the velocity on individual frequency-components U V  kdV=dk
is called (wave-dispersion).
The dispersion phenomenon leads to changing where, V, k, and dV/dk are average values for
of the shape of the wave train with travelled the range of frequencies making up the principal
distance. Each frequency component (that is, part of the pulse (Telford et al. 1990, p. 154).
each wave-phase) moves with its own individual When the phase velocity (V) increases with
velocity (the phase velocity, V). This is the increase of the component period, it is termed as
velocity with which a given point, marked on the (normal dispersion), and in this case the group
2.6 Effect of the Medium on Wave Energy 51

Fig. 2.32 A wave train LV LU


showing normal dispersion distance
as it is propagating. The
group velocity (U) and
phase velocity (v) are given
by the slopes of the lines
LU and LV respectively

time
0

velocity is less than the phase velocity (U < V). Dispersion phenomenon occurs in a disper-
For the opposite case (inverse dispersion), it is sive medium, as when surface waves are travel-
when phase velocity decreases with period we ling through a semi-innite medium which is
get (U > V). In the absence of dispersion, the overlain by a low velocity surface layer. Dis-
two velocities are equal (U = V) and no distor- persion of seismic body waves (P- and S-waves)
tion to the wav-form occurs. are too small to be detected in practice.
The Seismic Velocity
3

3.1 Introduction affected by these properties. The main factors


affecting the propagation velocity of seismic
The wave-propagation velocity plays a funda- waves are: rock lithology, elastic coefcients,
mental role in the theory and application of seismic bulk density, and fluid contents. These factors are
waves. It enters in the wave-motion equation as here-below briefly presented.
well as in the wave changes occurring at interfaces.
Seismic-energy partitioning at interfaces (reflec-
tion and transmission coefcients) is governed by 3.2.1 Lithology Effect
variations of the acoustic impedance which is, in
turn, dependent on the wave velocity. For a given In nature, rocks differ widely in their chemical and
seismic wave-type, velocity of any type of seismic physical properties. These properties (density,
waves is function of density and elastic constants porosity, fluid saturation, crystallization, mineral
of the medium in which the wave is moving. content, rock texture) make up the lithological
Further, we have for a given medium; each wave type of rocks. This means that the seismic velocity,
type has its own propagation velocity. Thus, for which is a function of the overall physical prop-
example, P-waves move faster than S-waves erties, depends on the lithological nature of the
which, in turn, move faster than Rayleigh waves. medium. Igneous rocks, for example, are normally
In the common earth material where Poissons characterized by their high seismic velocity com-
ratio is about (1/4), P-wave is nearly (1.7) times as pared with the sedimentary rocks.
fast as S-wave and (1.9) times as fast as Rayleigh It is very common that a rock type is charac-
wave in the same medium. That is why we see in a terized by a wide range of velocities rather than
typical earthquake seismogram, that P-waves by one single value. This is because of variation
arrive rst, followed by S-waves and then fol- of properties found within the one dened
lowed by Rayleigh waves (Fig. 3.1). rock-type. A sandstone rock, for example, may
have a velocity value anywhere in the range of
approximately (26) km/s. The corresponding
3.2 Factors Influencing Seismic range for limestone is (37) km/s. Usually, there
Velocity are overlaps in the velocity ranges for the different
lithologies. For this reason it is difcult to identify
Since velocity of a seismic wave is basically a the type of lithology on basis of velocity criterion
function of the physical properties (density and alone. The following Table 3.1 contains ranges of
elasticity) of the traversed medium, it is naturally values of P-wave velocity, bulk density, and

Springer International Publishing Switzerland 2017 53


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_3
54 3 The Seismic Velocity

Fig. 3.1 Long period 1 min


vertical-component
seismograms recorded at
Uppsala (Sweden).
(a) Skopje earthquake
a Skopje earthquake, July
26, 1963, mag. 6.0. b An
atmospheric nuclear
explosion, mag. 5.4.
Arrivals of P, S, and P wave S wave R wave
Rayleigh (R) waves are as
shown (Bth 1973) 1 min
(b) Atmospheric nuclear explosion

P wave S wave R wave

acoustic impedance for the most commonly coefcients increase in such a way that it offsets the
known rock-types (Al-Sadi 1980, p. 70). effect introduced by the density increase. An
empirical relationship between P-wave velocity
(v) and bulk density (q) for the common sedimen-
3.2.2 Elasticity and Density Effects tary rocks (sandstone, shale, limestone, anhydrite
) is given as follows (Gardner et al. 1974):
The wave motion equation of seismic waves
includes the velocity factor present in its mathe- qv kv1=4
matical expressions. For a solid homogeneous
medium, the propagation velocity for P- and where (k) is equal to 0.31 when (v) is in m/s and
S-waves (vp and vs) are functions of elastic constants equal to 0.23 when (v) is in ft/s.
of the medium (Lames elastic constants, k and l) as Density values for common rock types are
well as its bulk density (q). These functions are: found in Table 3.1 presented above.

vp k 2l=q1=2
3.2.3 Porosity and Saturation Fluid
and, Effects

vs l=q1=2 Porosity (dened as the pore volume per unit


volume) has an effective role in seismic velocities
Although the mathematical expressions of the because of its direct relation to the bulk density.
velocity-density relations show that velocity is In a porous rock, velocity is affected by porosity
inversely proportional to the square root of density, as well as on the type of interstitial fluid. In
it is a common observation that velocity appears to general there is an inverse relationship between
be increasing as density increases (Nafe and Drake porosity () and velocity (v) of rocks. Velocity of
1963; Gardner et al. 1974). The explanation for this a fluid-saturated rock is given by the following
discrepancy, is that as the material becomes more empirical formula which is normally called the
compact (that is as density increases) its elastic time-average equation (Wyllie et al. 1958):
3.2 Factors Influencing Seismic Velocity 55

Table 3.1 Propagation Medium Density, q Velocity, Vp Acoustic impedance, Z


velocities of compressional (gm/cm3) (m/s) (gm/s cm2)  104
waves, densities, and
acoustic impedances for Air 0.0013 330 0.004
selected media Aluminium (P) 2.70 63007100 170192
Anhydrite (P) 2.822.93 35005500 99161
Basalt (S) 2.703.30 55006300 149208
Chalk (P) 1.942.23 21004200 4194
Clays (P) 1.52.5 11002500 1763
Copper (P) 8.96 48205960 432534
Diabase (P) 2.803.11 58006600 162205
Dolomite (P) 2.752.85 35006900 96197
Dunite (S) 3.203.31 75008100 240268
Gabbro (P) 2.852.92 64506700 184196
Gneiss (P) 2.662.73 35007500 93205
Granite (P) 2.522.82 47506000 120169
Granodiorite (P) 2.672.78 46004880 123136
Gypsum (P) 2.312.33 20003500 4681
Ice (G) 0.971.07 31004200 3045
Limestones (P) 2.582.80 34007000 88196
Marble (P) 2.75 3750^6940 103191
Marl (G) 2.252.86 20003500 45100
Moraine (P) 1.52.0 10002700 1554
Oil (P) 0.60.9 1275 (23 C) 811
Peridotite (S) 3.153.28 78008400 246276
Rock salt (G) 2.142.18 42005500 90120
Sand (P) 1.601.90 6001850 1035
Sandstones (P) 2.152.70 21004500 45122
Shale (G) 2.412.81 27004800 65135
Water (G) 0.981.01 14301590 1416
Symbols (P, G, and S) indicate references:
(P) Parasnis (1971) (G) Gurvich (1972) (S) Sharma (1976)

1=v =vf 1  =vm involve the effects of the rock and fluid contents
individually. It has the form:
or, in terms of interval transit time, T (= 1/v), 1=v C1 C2

T Tf 1  Tm where C1 and C2 are constants, depending on


lithology and depth of burial.
where, (vf) and (vm) are the velocities of the fluid Velocity-change with porosity depends on the
content and rock matrix respectively, and (Tf) type of fluid (gas, oil, water) and the pressure it is
and (Tm) are the corresponding transit times subjected to. Velocity decreases with increase of
Another velocity-porosity relationship was gas saturation. Gas has more effect than liquid
developed (Pickett 1969) which does not explicitly (such as oil or water) in lowering velocity. It is
56 3 The Seismic Velocity

noted that only a small amount of gas, present in resulting in a velocity increase with both depth and
pores, produces sharp decrease in velocity, but geological age.
further increase in gas saturation produces only a Faust formula (Faust 1951), for a given geo-
minor effect. This phenomenon was applied as a logical interval, can be presented in terms of the
tool for detection of hydrocarbon accumulation interval transit time, T (=1/v). This form is pre-
in oil traps. Being more compressible than oil or sented as follows (Pennebaker 1968):
water, gas has the effect of lowering the velocity
much more than the presence of oil or water as Th Chn
interstitial fluids. Due to its effect on the elastic
properties of rocks, even a small quantity of gas or,
present in the pores, would result in large
log Th n  log h log C
reduction in velocity (Sheriff and Geldart 1995,
p. 109). Clearly, this property has an important
where (C) is constant related to the rock lithol-
application in hydrocarbon exploration.
ogy, pore pressure, and geological age.
For sand-shale sequence, Pennebaker (1968)
found that the index (n) has a value of about
3.2.4 Depth and Geological Age
(1/4). This is considered to be representing a
Effects
geological section existing under normal
compaction-pressure. As it is seen above, the
With depth of burial and with geological age, in
(T-h) relation is linear when it is expressed in the
general, rock material becomes more compact.
log-domain. Thus, when a linear (log T log h)
This would result in increase in elastic constants
relation is obtained in a certain study, the geo-
as well as in the density and, consequently,
logical section is considered to be under normal
increase in velocity. Dependence of velocity (v),
compaction pressure conditions. Deviation from
of a particular geological bed, on geological age
the linear relation (straight-line relation) is nor-
(g), existing at depth (h), is expressed by the
mally considered as indicative of abnormal
following empirical formula (Faust 1951):
pressure situation. This is another useful and
effective tool that can be used to investigate
vg; h kg  hn
subsurface pressure conditions and as an indica-
where (k) is constant (equal to 125.3 when depth tor for other stratigraphic features.
in feet, velocity is in ft/s, and geological age in
years). The index (n) is found in this study to be
equal to (1/6). In a later study (Gregory 1977) 3.2.5 Overburden Pressure Effect
made on sand- and shale-formations data, the
index (n) is found to be equal to (1/4). A theoretical relationship developed by Gassman
Variation of velocity with the depth of burial (1951) showed that velocity of a rock-bed model
was investigated by a number of geophysicists made up of spherical tightly-packed grains varies
(e.g. West 1950; Kaufman 1953; Acheson 1963), as (P1/6), where P is the applied compaction
conrming the general trend of velocity increase pressure. The similarity of this result to Fausts
with depth. empirical formula (Faust 1951) suggests that
As the function, v(h, g) is showing, geological increase in velocity (i.e. decrease in the interval
age bears a relationship with velocity similar to transit time) is due to increasing pressure
that of the depth factor. It shows that velocity is in imposed by the overburden rocks.
general increasing with depth and geological age. Compaction of a saturated porous layer is
The explanation is that, grain compaction increa- function of the weight of the overlying material (i.e.
ses with increase of depth and geological age. In due to the overburden pressure, Po) and the pressure
consequence, elastic constants are increased of the saturation fluid (the pore pressure, Pp). Since
3.2 Factors Influencing Seismic Velocity 57

the effect of (Pp) in compressing the porous rocks is phenomenon. In case of anisotropic medium,
in opposite direction to that created by (Po), the velocity varies with changes in propagation
compaction net effect (compaction pressure, Pc) is direction. The main factors affecting velocity can
proportional to the pressure difference (Po Pp). be summarized as follows:
That is:
(i) Lithology
Pc Po Pp In general, igneous and metamorphic rocks
have larger velocity values than sedimen-
As rock compaction process brings about a tary rocks. Sub-types of rock lithologies
corresponding change in the physical properties, it are characterized by wide ranges of
is expected that seismic velocity (or interval transit velocities which are overlapping with each
time) would vary with the applied compaction other. For example, the velocity-range for
pressure. Velocity increases with increasing sandstone (25) km/s is overlapping with
compaction pressure and decreases with increase that of shale (1.54.5) km/s, and so on.
of interstitial pore pressure. This velocity-pressure (ii) Elasticity and Density
relationship forms the basis used in predicting Theoretically, velocity is directly propor-
subsurface pressure conditions from seismic tional to elastic constants and inversely
velocity data. Use of seismic velocity as an indi- with density. However, because of the
cator of pressure conditions was used by several greater effect of elasticity, it is observed
workers like Pennebaker (1968), Reynolds (1970), that velocity has an apparent direct pro-
Louden et al. (1971), and Aud (1974). Departure portionality with density changes.
of the measured velocity of a given geological (iii) Porosity and Saturation Fluid
section from the normal-compaction trend serves Velocity is inversely proportional to
as indicator of abnormal-pressure zone. porosity, and the change in velocity
depends on the type of pore fluid. Velocity
is lowered when the pores are lled with
3.2.6 Other Velocity-Affecting water or oil, and more lowered when the
Factors pores are lled with gas.
(iv) Depth and Geological Age
In addition to the above mentioned factors, there In general, velocity increases as both of
are other less important factors which can cause the overburden depth and geological age
small effects on seismic velocity. Examples of increase. A power-law relationship is
such factors are pore shape, anisotropy, tempera- connecting the velocity with the depth and
ture, and wave-frequency. It is found that velocity geological age. The power index for
decreases slightly with increase of temperature. It sand-shale sequence is found to be around
is found that, for an increase of 100 C, velocity the value of (1/61/4).
decreases by about 5 % and that the velocity (v) Overburden Pressure
drop-rate is considerably more when the rocks are Due to compaction caused by the overbur-
saturated with heavy crude oil or tar. Velocity in den weight, elastic constants as well as
water saturated rocks experiences sudden increase density get increased. There is a direct rela-
as temperature is lowered to the freezing point tion between velocity and the net com-
(Sheriff and Geldart 1995, pp. 120121). paction pressure which is equal to the
Velocity is practically unchanged over a broad difference between pore pressure and over-
frequency range. Small change is observed in case burden pressure (geostatic pressure as it is
of body-wave dispersion (velocity variation with sometimes called). Velocity increases with
frequency). Low-frequency components are fas- increasing compaction pressure and decrea-
ter because of frequency-dependant absorption ses with increase of interstitial pore pressure.
58 3 The Seismic Velocity

Fig. 3.2 Main factors vp km/sec


affecting seismic P-wave vp vp vp
velocity (vp). The symbols 8
SH, SS, LS, and AN denote
shale, sandstone, limestone, 6 AN
LS
and anhydrite respectively 4 SS
SH
2 depth
lithology density porosity overburden pressure

vp vp vp vp
water saturated

oil
gas
oil saturated
Pore pressure fluid saturation temperature frequency

(vi) Other Minor Factors express the propagation velocity (V) as a math-
Pore shape: Velocity decreases when ematical function of the travel-distance, V(x) or
pores are elongate. function of travel-time, V(t). For practical
Anisotropy: Velocity varies with changes applications, velocity is, more often, expressed as
in propagation direction. function of vertical reflection time, that is, as
Temperature: Velocity slightly decreases function of the two-way (reflection) vertical time
with increase of temperature. V(T). In a homogeneous, uniformly changing, or
Frequency: Velocity practically unchan- layered medium, the distance-versus-time func-
ged with frequency. With normal disper- tions can be linear, curved, or segmented
sion, low-frequency components are respectively. The corresponding V(T) functions
faster because of frequency-dependant can be plotted as curved or straight lines. Typical
absorption phenomenon. velocity functions are shown in Fig. 3.3.
Due to the geological complexity of the earth,
Main factors affecting velocity are shown in it is not, in practice, possible to express seismic
Fig. 3.2. velocity-variation in the form of an explicit
mathematical function. However, under certain
conditions, a geological medium is approximated
3.3 The Velocity Function by a dened set of specications that can allow
derivation of a mathematical function describing
Seismic propagation velocity, as we have seen the variation of velocity along the travel-path.
from the previous discussion, depends on a One of these models is one in which the instan-
number of variables, like type of lithology, geo- taneous velocity (V) increases linearly with depth
logical age, density, porosity, pressure, and depth. (h). This model is expressed by the linear func-
In seismic reflection exploration work, velocity is tion; V(h) = V0 + kh, where (V0) is the velocity
usually presented as mathematical functions. The value at the surface (h = 0) and (k) is the
velocity function is taken to describe its variation velocity-depth gradient. The signicance of this
with reflection travel-distance (or, more often, relation is that it gives a close approximation to
with travel-time). the actual velocity variation observed in many
Since the real earth is not homogeneous sedimentary-basin areas (Dobrin 1960, p. 77).
(typically layered medium), seismic waves move Derivation of the ray-path geometry (which turns
with different velocities in different parts of its out to be of circular form) and travel-time func-
travel-path. In this case, it becomes very useful to tions, are given by several authors (see for
3.3 The Velocity Function 59

Fig. 3.3 Sketches of X(T) V(T)


velocity functions for three
cases: a homogeneous
medium, V(T) is line of
zero slope, b uniformly A
changing velocity, V(T) is
line of constant slope,
c layered medium, curve is B
made up of zero-slope
linear segments T T
A B C
C

example, Nettleton 1940, pp. 257261 or Sheriff of the location in the travel-path of the moving
and Geldart 1995, pp. 9394). wave front. This means that at any point of the
travel-path, the velocity can be measured or
computed at that point. This is the (instantaneous
velocity) which can be dened to be the velocity
3.4 Types of Velocity Functions
of wave motion at a given point in a medium
traversed by an advancing seismic wave.
Unlike electromagnetic waves which are all of
Mathematically, the instantaneous velocity (v) at
one type (e.g. wave of visible light which is of
a point is given by the slope of the tangent to the
S-wave type), seismic waves are of many types,
distance-versus-time curve at that point (Fig. 3.4).
each of which has its own velocity-value which
The instantaneous velocity, (v) of a wave
is dependent upon the density and elastic con-
moving along a distance (x) is dened as:
stants, as mentioned above. Because of the
inhomogeneous nature of the crustal part of the v dx=dt
Earth (which is typically made up of layered rock
media), several types of velocity functions are
needed to express velocity variation as function
of travelled distance. Five velocity-types are in
common use in seismic exploration. These are:
(i) Instantaneous Velocity (V) v = dx/dt
(ii) Interval Velocity (IV)
(iii) Average Velocity (AV)
(iv) Root Mean Square (RMS) Velocity (RV)
x(t)
(v) Stacking Velocity (SV)

3.4.1 Instantaneous Velocity

In cases where the subsurface geology is of t1 t


variable nature, the seismic velocity which is
Fig. 3.4 Denition of the instantaneous velocity (v),
governed by the physical properties of the med- given by the slope of the tangent to the xt curve,
ium will accordingly be changing with changes v = dx/dt at t = t1
60 3 The Seismic Velocity

The closest example to the instantaneous covering that distance. In seismic exploration
velocity is the velocity function in which the work, the average velocity is computed for a set
interval transit time (which is equal to the of layers, usually starting from the datum plane
reciprocal of velocity) is measured across each down to the required reflector level.
depth-interval of one meter down a drill hole. For a vertical travel-path, the average velocity
The resulting record (the sonic log) gives detailed of a geological section made up of (n) layers will
velocity-information at points spaced by 1 m be given by dividing the total thickness hn (=
interval. This is a practical approximation of the Dh1 + Dh2 + Dh3 +  + Dhn) by the travel-time
instantaneous velocity-variation with depth. tn (= Dt1 + Dt2 + Dt3 +  + Dtn), that is:
AVn hn =tn

3.4.2 Interval Velocity or,


Dh1 Dh2 Dh3    Dhn
The interval velocity (IV) is dened to be the AVn
average velocity over an interval of the Dt1 Dt2 Dt3    Dtn
travel-path. It is usually measured or computed
for individual geological layers. Thus, the inter- An example of computation is shown in
val velocity, of a geological formation of thick- Fig. 3.6.
nesses (h) and interval transit time (t), is The average velocity (AV) can be calculated
given by IV = h/t. For a multi-layer geolog- for a multi-layer section, given the interval
ical section, the interval-velocity function is velocities (IV) for each of the section layers. This
can be done by the following formula:

AVn = (IVn . tn) / tn

constructed by computing the interval velocity


for each layer of that section (Fig. 3.5). It is also possible to derive the interval
velocity of a certain layer in a pack of layers,
given the average velocity data. Thus, in order to
3.4.3 Average Velocity
compute the interval velocity (IVn) of the nth
layer, we need to know the two average veloci-
As the name implies, the average velocity (AV)
ties, (AVn) and (VAn1) for the section from the
is obtained from dividing the total distance
datum level down to the base and down to the top
travelled by the wave by the time spent in

Fig. 3.5 Interval velocity


function of a three-layer
geological section. IV1 at Layer-1 h1 , t1 IV1 = h1 /t1
depth h1, IV2 at depth
h1 + h2, and IV3 at IV2 = h2 /t2
depth h1 + h2 + h3 Layer-2 h2 , t2

Layer-3 h3 , t3 IV3 = h3 /t3

depth, h
3.4 Types of Velocity Functions 61

Fig. 3.6 Average velocity datum plane


(AV1, AV2, AV3) function of
one, two, and three-layers
of thicknesses (Dh1, Dh2, Layer-1 h1 , t1 AV1 = h1 /t1
Dh3). The corresponding
interval travel times are
(Dt1, Dt2, Dt3) Layer-2 h2 , t2 AV2 = (h1+h2)/(t1+t2)

Layer-3 h3 , t3 AV3 =(h1+h2+h3)/(t1+t2+t3)

depth, h

of that layer respectively. Using the corre- medium (medium made up of horizontal layers of
sponding travel times, (Tn) and (Tn1), the different interval velocities). A three-layer model is
required formula is: shown in the following Fig. 3.7.

IVn = (AVn .Tn - AVn-1 . Tn-1) /( Tn - Tn-1)

3.4.4 Root Mean Square Velocity The root mean square velocity (RV) can be
calculated from the interval velocity data (IV),
Root mean square (RMS) velocity (RV) is dened down to the nth layer, by use of the following
as the square root of the average of the squares of the formula:

RVn = [(IV2n . tn) / tn]1/2

weighted interval-velocities, where the weighting From the RMS velocity (RV), it is possible to
factors are the layer thicknesses or the interval transit derive the interval velocity (IV), given the two
times. In seismic exploration, the RMS velocity, like RMS velocities for the section, from the datum
other types of velocities, is usually computed for level down to the base (RVn) and down to the
vertical ray paths of waves traversing a multi-layer top (RVn1) of that layer. Using the

Fig. 3.7 Root mean


square (RMS) velocity Layer-1
(RV1, RV2, RV3) function of h1 , t1
RV1 = IV1
one, two, and three-layers
of thicknesses (Dh1, Dh2,
Dh3). The corresponding Layer-2 RV2 = [(IV21 t1 + IV22 t2)/(t1 + t2)]1/2
interval travel times are h2 , t2
(Dt1, Dt2, Dt3)

Layer-3
RV3 = [(IV21 t1+IV22 t2+IV23 t3)/(t1+t2+t3)]1/2
h3 , t3

depth, h
62 3 The Seismic Velocity

corresponding travel times, (Tn) and (Tn1), the gather-traces. This is one of the fundamental
required formula is (Dix 1955): processing steps (called NMO correction) which

IVn2 = [(RVn2 . Tn RVn-12 . Tn-1)/(Tn - Tn-1)

For a given geological section, the RMS velocity bring about coherency of the reflection arrivals
is typically a few percents larger than the corre- which, on stacking, give enhanced reflection
sponding average velocity (Sheriff 1973, p. 228). event. It is called stacking velocity because of its
role in enhancing the stacked reflection signal.
Since its direct role is in the NMO correction, the
3.4.5 Stacking Velocity term (NMO velocity) will be more appropriate
than the commonly applied term (stacking
Stacking velocity is the main velocity variable velocity). The CMP concept is explained in
that enters in the NMO-correction formula (ex- Sect. 4.4.
plained in Sect. 4.3). It is applied to remove the Role of the stacking velocity in enhancing the
time-contribution of the receiver-to-source dis- reflection signal is schematically shown in
tance (called, offset) from the total reflection Fig. 3.9.
travel-time. This is claried in the following In dipping parallel reflectors (dip angle, h),
(Fig. 3.8). with parallel velocity layering, the stacking
Stacking velocity is determined by velocity velocity (calle it VS) is equal to RMS velocity
analysis technique whereby the time contribution (VRMS) divided by cosine of the dip angle (h)
of offset is removed before stacking of the CMP that is (Sheriff and Geldart 1995, p. 134):

Fig. 3.8 Denition of the


stacking velocity. It enters
trace at trace at
S offset , x
in the NMO correction R x=0 x>0
formula (T = Tx T0),
where T0 and Tx are the
reflection travel-times for
zero-offset and x-offset
respectively h T0
T
reflector Tx
RP RP

T0 = 2h/V
Tx = (4h2 + x2 )1/2 / V

T = Tx - T0 = [(x/ V)2 + (T0)2]1/2 -T0

V is the stacking velocity


3.4 Types of Velocity Functions 63

CDP-gather stack traces other than the true propagation direction. In seismic
A B C exploration, the apparent velocity is often dealt
0 x
with in connection with the movement of plane
wave-front advancing along a path inclined with
respect to the ground surface. A plane wave
A approaching a ground surface with true velocity
(V), along a ray-path making an angle (h) with the
B normal to the surface will have an apparent velocity
(Va) of its motion along the surface (Fig. 3.10).
During a time interval (t), the wave front
C moves a distance of (Vt) with the true velocity
(V), while, at the same time, the moved distance
T0 Tx at the surface (x) is covered by the apparent
velocity (Va), that is:
Fig. 3.9 Role of the stacking velocity in the NMO-
correction of a CDP trace gather. In this model, three
stacking velocities were applied (cases: A, B, and C). The Va Dx=Dt
optimum velocity (case, B) gave the highest staked signal
hence,
VS VRMS =cos h Va V=sin h

This relation between the stacking velocity This relation shows that the apparent velocity
and RMS velocity can be applied by interpreters is always greater than the true velocity by a
in their interpretation work of seismic reflection factor depending on the angle of approach (h, in
data and in velocity-changes studies. this example). The apparent velocity approaches
Stacking velocity is sometimes called RMS innity when (h = 0), that is when the wave path
velocity, because stacking velocity is the nearest direction is perpendicular to the surface plane. In
in value to RMS velocity in a multi-layer med- any case the apparent velocity is always greater
ium. Like RMS velocity, stacking velocity is than the true velocity with which the wave is
slightly greater than the average velocity. approaching the horizontal surface plane.
Another point of interest which is related to
this subject is the apparent wavelength. Suppose
3.4.6 The Apparent Velocity that the measurements were taken for time of one
period (t = s), distance measured over the sur-
The Apparent Velocity (Va) is dened as the face corresponding to time of one period will be
propagation velocity measured in the direction

x
Va t

Wave Vt
wave motion
front
direction

Fig. 3.10 Ray path geometry of a plane wave approaching the surface plane at an approach angle (h). Velocity in the
ray-path direction (V), and velocity of wave front measured on the horizontal plane surface is the apparent velocity (Va)
64 3 The Seismic Velocity

one apparent wavelength (x = ka), hence (using 3.4.8 Representations of the Velocity
the relationship; k = Vs): Functions

ka k=sin h 2p=ka The velocity types that play important roles in


seismic exploration are: interval velocity, aver-
where (ka) is the corresponding apparent wave age velocity, RMS velocity, and stacking veloc-
number. ity. In reflection seismology these velocities are
usually plotted as functions of reflection travel
time, as shown in Fig. 3.11.
3.4.7 The Group and Phase Velocities As far as velocity types are concerned, there
are other types of velocities of less importance to
The group velocity and phase velocity are dis- the seismic exploration applications. Most
tinguished in cases of dispersive waves. This important of these are the apparent velocity,
phenomenon has practically no applications in group velocity, phase velocity.
the eld of seismic-exploration, since the body
waves, which are used in seismic exploration
show no signicant dispersion. Wave dispersion 3.5 Velocity Determination
is common with surface waves under certain Methods
conditions. Wave dispersion occurs as result of
variation of velocity with frequency, and in this Seismic velocity plays an important role in all
case, two velocities are distinguished; the group activities involved in seismic exploration survey-
velocity and phase velocity. ing. Its importance stems from the fact that the end
The group velocity is dened to be the velocity product of any seismic survey is a time-image (and
with which the seismic energy-packet (repre- not depth image) of the subsurface geology. Thus,
sented by the wave-train envelope) travels. The to convert the time-image data into depth domain,
phase velocity, on the other hand, is the velocity velocity must be made available. In seismic
of a certain frequency component of the moving reflection surveying, velocity computations, aim
wave. As usual, the two velocities differ in value at nding the seismic velocity (average- or
and consequently, a wave-peak, or a wave-trough interval-velocity) expressed as function of depth.
appears to move within the wave-train. Detailed There are several ways to compute the seismic
account on this subject is found in the geophysi- velocity-function. Velocity computations are
cal literature as in Richter (1958, pp. 243244), made either by methods based on borehole data or
Telford et al. (1990, pp. 153154), and Sheriff by methods based on analysis of seismic data. The
and Geldart (1995, pp. 6062). two groups of velocity determination methods are:

Fig. 3.11 Representation seismic velocity


of the main types of
velocity functions used in
seismic reflection
average velocity
exploration
RMS velocity or
reflection stacking velocity
time
interval velocity
3.5 Velocity Determination Methods 65

Fig. 3.12 Method of well


velocity surveying. The
Seismic energy source average velocity
detector is placed at a
geological-formation
boundary. The curve is
representing the computed
average velocity as
function of depth
detector

depth

Borehole-Based Methods wall. It may be a hydrophone-type detector sus-


pended inside the well which is lled with the
(i) Well Velocity Surveying
drilling fluid.
(ii) Up-hole Velocity Survey
The average velocity (Vav) is computed from
(iii) Continuous Velocity Survey
the slant travel time (T) of the direct wave
Seismic Data Analysis Methods recorded by the detector placed at depth (h) using
(iv) (X2 T2 Method) the formula (Fig. 3.13):
(v) (T T) Method
Vav h=Ts cos h
(vi) Velocity Analysis
(vii) Seismic Inversion The interval velocity of a certain geological
formation is derived by dividing the formation
thickness by the interval transit time of that for-
3.5.1 The Well Velocity Survey mation. Another way of computing of the inter-
val velocity is calculated by the mathematical
The normal well velocity surveying proceeds by relationship connecting the interval velocity with
generating seismic waves from a seismic energy the average velocity.
source located on the surface near the well head. The computed velocity function is very
The directly arriving wave is recorded by a important for the interpretation process. It is used
detector placed at a certain depth inside the well for calibrating the sonic log and check the inte-
(usually at a boundary of a geological formation). grated time of the sonic log, hence the name
The shooting and recording process is repeated at (check-shot surveying) which is sometimes used
all of the geological-boundaries penetrated by the as another name for the well shooting method.
drill-hole. For more detailed surveying, record-
ings are made at additional intermediate Vav = h / Ts cos
detector-positions of smaller spacing. From the surface
source-to-detector travel time, corrected to the
vertical path, the average velocities are calculated
and plotted against depth (Fig. 3.12). h T Ts
The source is either dynamite charge red in a Th = Ts cos

shallow drill-hole or air-gun operated in a mud Vav = h / Th
pit. The detector is normally a specially-designed
geophone provided with a lever that makes the
geophone to be well-pressed against the borehole Fig. 3.13 Computation principle of the average velocity
66 3 The Seismic Velocity

3.5.2 The Up-Hole Velocity Survey The electronic structure of the sonde is
designed in such a way that the output is made to
This method is applied to determine velocity be the difference in the travel-times to the two
changes in relatively shallow depths, in the range receivers. The time difference, measured in
50100 m. It typical application is velocity- time-units per one-foot, called (interval transit
determination of the near-surface weathered zone. time), is plotted (normally in micro-second units)
The same principles used in well-velocity survey- against depth to give the continuous wiggly
ing are applied in this method, but with the curve known as the (sonic log). Sonde basic
source-detector conguration reversed. In the structure is shown in (Fig. 3.14a).
up-hole case, the sources (small dynamite charges) To avoid tilting and hole-irregularities effects,
are placed inside the borehole at few-meter spacing two seismic-pulse sources are used instead of one,
and sequentially red. The detection system is making what is called a (borehole-compensated
placed on the surface at a location near the sonde) as shown in (Fig. 3.14b). Seismic pulses
well-head. Velocity computation follows the same are emitted alternately from the two sources and
equations quoted above (Fig. 3.13). the transit times from the two oppositely traveling
refracted P-waves are averaged electronically.
The borehole compensated sonde (BHC) gives
an average interval transit time which is plotted
3.5.3 Continuous Velocity Logging
on a paper strip. The produced log in this case is
normally referred to as BHC sonic log.
This method uses a recording system consisting
The BHC-log is used in computing synthetic
of a borehole-logging tool, called the (sonde). In
seismograms, in identifying lithologies, and in
its standard form, it contains a seismic-pulse
determining formation boundaries (Fig. 3.15).
source and two receivers, one foot apart, and
The interval transit time can be integrated down
source placed three feet from the nearest receiver.
the well to give the total travel time of the part of
Sound pulses, emitted from the source at uniform
the well for which continuous velocity logging
time intervals, are detected by the two receivers.
was conducted.
Due to the fact that P-wave velocity in the dril-
ling fluid is lower than that of the rock-medium Seismic Data Analysis Methods
surrounding the well, the transmitted P-wave gets Reflection travel-time function, in a homoge-
refracted at the wall-side, moving into the rock neous and isotropic medium, is of a hyperbolic
medium and then recorded by the receivers.

sonic log
(a) (b) geological column transit time, sec
140 90 40

shale
S S

R limestone
R R
R

non-compensated compensated sandstone


single-source sonde double-source sonde

Fig. 3.14 Schematic representation of the


source-receiver conguration used in the logging sonde.
a None-compensated, single-source sonde. b Compen- Fig. 3.15 Use of BHC sonic-log in determination of
sated, dual-source sonde formation boundaries and in recognizing lithologies
3.5 Velocity Determination Methods 67

form. The reflection time (Tx) is function of h i h i


V2 x1 2  x2 2 = T1 2  T2 2
reflector depth (h), receiver-offset (x), and prop-
agation velocity (V). The well-known travel-time
function for horizontal reflector is: To get more accurate results, the computation
is repeated for larger number of offsets.
h i1=2
Tx x=V2 T0 2 ;
where; T0 2h=V:
3.5.5 (T T) Method
Dependence of the reflection travel-time on
velocity can be used as basis for analytical determi- Another approach, based on the travel-time func-
nation of seismic velocities. Two main approaches tion is the (T T) method. By denition, (T)
for velocity determination belong to the analytical stands for the difference between reflection
methods. These are the reflection travel-time com- travel-time (Tx) of a seismic wave received by an
putations and velocity processing analysis. x-offset detector and the corresponding zero-offset
time (T0). It is expressed by:

3.5.4 (X2 T2 Method) h i1=2


DT x=V2 T20 T0 ; exact form
Since the early years of seismic reflection explo- DT x2 =2T0 V2 ; approximate form
ration, methods based on travel-time analysis were
applied to determine velocity. In 1938, Green These equations show that (T) is function of
(1938) published a simple method of velocity cal- the three variables (x, T0, and V), implying that it
culation based on travel-times of reflection arrivals is possible to calculate the velocity (V), given
using the reflection travel-time equation, (Tx)2 = ( (T, x, and T0). The parameter (T) can be
x/V)2 + (2 h/V)2 or (Tx)2 = (x/V)2 + T20, and readily measured from the reflection record with
computing the velocity from the slope given by (1/ reasonable accuracy. Since the other variables
V)2 of the (Tx)2 (x)2 straight-line plot. The (x and T0) are known quantities (i.e. they can be
symbol (T0) represents the two-way vertical determined from a shot record), the velocity
reflection time from a reflector at depth (h). Fur- (V) can be computed. To increase the accuracy,
thermore, the depth (h) of the reflector, can be cal- many shot records should be used in the com-
culated from the intercept (T0)2. putations. One must remember that these equa-
Another way of using the reflection travel-time tions are all based on the assumption that the
equation is by recording reflection arrivals at two reflectors are horizontal planes. If, however, the
different-offset receivers. Thus, for such two reflectors are dipping, the error can be minimized
reflection experiments, the times (T1) and (T2) by using a centre-spread and using the average
corresponding to offsets (x1) and (x2) are given by: value of (T) by measuring it from traces having
h i the same offset and located on either side of the
T1 2 x1 =V2 T0 2 source (Fig. 3.16).
h i It should be noted here that both of these two
T2 2 x2 =V2 T0 2 methods (X2 T2 and T T methods) are
now obsolete and they are replaced mainly by the
hence, the velocity (V) can be calculated from: velocity analysis techniques.
68 3 The Seismic Velocity

Fig. 3.16 Shot record


showing three
reflection-events. The
parameters (T0 and T) are
shown for the rst reflector

T0

T T

3.5.6 Velocity Analysis Method NMO-corrected. The correct stacking velocity is


the velocity which, on NMO correction, makes all
This method is closely related to the (T T) reflection events in the gather in phase, and when
method described above. Both methods depend these are stacked, a strong reflection signal is
on computing (T) as accurately as possible. The obtained. Application of too-low velocity in the
process of velocity analysis mormally follows a correction formula results in NMO-
trial-and-error approach. A set of trial velocity overcorrection, and application of too-high
functions are applied in NMO-correcting the velocity will result in NMO-under-correction.
traces of a CMP-gather. The criterion used in Deviation of the applied velocity from the opti-
recognizing the correct velocity is the S/N ratio mum value (that is the velocity is not too low and
of the stacked NMO-corrected reflection signal. not too high), shall lead to fall of the amplitude of
The end- result of the analysis is a stacking the stacked reflection signal. This phenomenon is
velocity for each reflection event, plotted as shown schematically in Fig. 3.9.
function of reflection two-way vertical time. Velocity analysis is one of the mandatory
The stack trace is obtained from summing steps normally executed in processing of the
(stacking) the CMP-gather traces after being seismic reflection data. The procedure followed
3.5 Velocity Determination Methods 69

in the analysis is explained in more details in Since density variation is very small, com-
Sect. 10.7.1. pared with velocity variation, that is putting
(q1 = q2), the direct (A-V) relationship can be
3.5.7 Seismic Velocity Inversion obtained which is:

The amplitude of a reflection signal (A) is V2 1 A=1  A  V1


dependent on the reflection coefcient (R) which
is, in turn, depending on the acoustic impedance This is an inverse problem in which the
(Z) which is equal to density (q) multiplied by acoustic impedance (expressed by velocity) can
velocity (V). This relationship (Z = qV) shows be obtained from amplitude data. If the velocity
that velocity can be determined from amplitude (V1) of the surface layer of a layered medium is
measurements, given the seismic impedance and known then using the inversion formula,
density data. The basic principle, upon which the V2 = [(1 + A)/(1 A)]  V1, the velocity (V3)
seismic inversion is based, will be presented in a of the neighboring deeper layer is computed. By
simplied way, as follows: repeating the computation, velocities of the rest
For a plane seismic wave perpendicularly of layers are sequentially determined.
incident at an interface separating two layers of Normally this approach is applied in trans-
acoustic impedances (Z1 and Z2), the reflection forming a seismic stack section into acoustic
coefcient (R) is given by: impedance section, or into what is called
pseudo-impedance when density is ignored. The
computations are normally carried out by soft-
R Z2 Z1 =Z2 Z1 ware especially designed for this purpose.
q2 V2 q1 V1 =q2 V2 q1 V1

Assuming the amplitude of the incident wave


3.6 Uses of the Seismic Velocity
is unity (1), and the reflected amplitude is (A),
Data
the reflection coefcient (R), by denition,
becomes equal to (A) and hence, we can use the
All types of the seismic velocity have important
following equivalent form:
role in the seismic exploration activities (data
acquisition, processing, and interpretation).
A Z2 Z1 =Z2 Z1
Velocity enters in the travel-time functions of all
q2 V2 q1 V1 =q2 V2 q1 V1
body and surface waves (direct, reflected, refrac-
giving: ted, and diffracted waves). To start with, velocity
governs reflection and transmission coefcients.
Z2 1 A=1  A  Z1 In processing of seismic data, velocity forms an
important factor in travel-time corrections like
70 3 The Seismic Velocity

Table 3.2 Fields of application and precision assessment of various types of velocity (after Al-Chalabi 1979)
Velocity Main uses Precision requirements
Stacking velocity Stacking of seismic sections Modest to low
Migration processing Modest to low
Estimation of RMS velocity Dependant on situation
RMS velocity Migration-velocity estimation Generally modest
Interval-velocity estimation Dependant on situation
Average-velocity estimation Dependant on situation
Interval velocity Lithologic and stratigraphic studies High to modest
Interpretation works Modest to low
Abnormal pressure detection High to modest
Ray tracing computations Dependant on situation
Migration processing Generally modest
Average-velocity estimation Dependant on situation
Average velocity Depth conversion Generally modest
Interpretation works Modest to low
Precision requirements: high = 0.11.0 %, modest = 15 %, low >5 %

static and dynamic (NMO) corrections. Amplitude conversion and in mapping structural and strati-
compensations (as in geometrical spreading and graphic features.
inelastic absorption) and seismic migration Summary of elds of application and preci-
depend on velocity. In interpretation activities, sion assessment of the various velocity-types are
velocity has a fundamental role in time-to-depth given in the following Table 3.2.
Seismic Wave Reflection
and Diffraction 4

In the real Earth Crust, a seismic wave may meet A diffraction event bears a marked relation
a variety of geological changes. Typically, the with reflection events since both are types of
media traversed by seismic waves are made up of seismic energy generated from an intervening
layered rock formations of different physical reflector. A plane surface (surface-reflector)
properties and different geometrical shapes. In causes reflection and point obstacle (point-
such environments, some of the seismic wave reflector) causes diffraction.
energy gets reflected from interfaces or diffracted
from structural obstacles. The rest of the incident
wave is transmitted through the interfaces with
their ray-paths being bent (refracted) in case of 4.2 Wave Changes at Reflection
inclined incidence and with no bending when Interface
ray-path is perpendicular to an interface.
In this chapter, discussion shall deal with the The reflection process involves two main types
reflection and diffraction of seismic waves since of changes. These are: change in the propagation
they are physically more closely related to each direction and change in the energy content
other. Transmission with its special case, the (Fig. 4.2).
refracted transmission (refraction) shall be dealt The two types of changes occurring in the
with separately in the following chapter. reflection phenomenon involve the wave energy
content and its travel-path geometry. These are:

(i) Change in Energy content:


4.1 The Commonly-Recorded In the process of reflection, the energy con-
Seismic Events tent of the incident wave is shared among all
of the reflected and transmitted waves.
A seismic event on a shot record may be created Consequently, the amplitude of any of the
as result of certain type of wave arrival (direct, reflected waves is always less than that of the
refracted, reflected, or diffracted) depending on incident wave. Distribution of the incident
the nature of the involved interface and on the seismic energy is governed by a measure-
detector position. All of these four types of ment parameter, the reflection coefcient.
seismic events play important roles in seismic (ii) Change in Propagation Direction:
exploration activities. In normal seismic reflec- At the interface, part of the incident seismic
tion records, these events are diagnosed by their energy is reflected following a travel path
recorded wave-arrivals as shown in Fig. 4.1. dened by the law of reflection which states

Springer International Publishing Switzerland 2017 71


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_4
72 4 Seismic Wave Reflection and Diffraction

diffraction source reflection direct refraction


arrival point arrival arrival arrival

subsurface interface Layer-1


Point Layer -2
diffractor

Fig. 4.1 The four principal types of wave arrivals (direct, reflection, refraction, and diffraction) representing the
commonly recorded seismic events

S R These two types of changes are dealt with in


some details in the following discussions.

4.2.1 Reflection Coefficient


at Inclined Incidence

i r A P-wave hitting an interface at an angle of more


than zero degrees with the normal (inclined
incidence) will lead to two reflected waves
RP (P and SV) and two transmitted waves (P and
SV). The law of wave reflection at interfaces
Fig. 4.2 Reflection of a seismic wave from an interface. (angle of incidence and that of reflection are
Travel path from the source point (S) to receiver (R) via
the reflection point (RP). Change in travel-path direction equal) is applicable only for alike wave-phases.
and reduction in energy content Thus, when the incident wave is a P-wave, the
reflected P-wave follows the normal law of
that angles of incidence and reflection are reflection (Fig. 4.3).
equal provided that these are of the same Since the energy of the incident wave is
wave types. The geometry of the reflection shared by the four generated waves (reflected
travel-paths are normally expressed by cer- P and SV and transmitted P and SV), any of the
tain adequate mathematical functions. generated waves will be of less energy level (less

Fig. 4.3 Inclined P SV


incidence of P-wave at an
interface. Two reflected and P
two transmitted (P and SV)
waves
Z1 = 1 V1
M1, V1, 1
M2, V2, 2 Z2 = 2 V2
P

SV

incidence angle = reflection angle =


4.2 Wave Changes at Reflection Interface 73

amplitude) compared with the incident wave. Sometimes, the reflection coefcient is dened
The ratio of the reflected amplitude to that of the in terms of energy ratios instead of amplitude
incident represents the reflection coefcient. ratios. With this approach, the coefcients are
Zoeppritz equations provide the mathematical expressed by squares of the amplitudes.
expressions for the reflection and transmission By applying the stress and strain continuity-
coefcients of all of the produced waves (two conditions at the interface level (z = 0), it is
P-waves and two SV-waves in this example). possible to derive expressions for the reflection
The important feature here is that the reflection and transmission coefcients in terms of the
and transmission coefcients are functions of contrast in acoustic impedance (see Sheriff and
both of the angle of incidence and the contrast in Geldartl 1995, p. 76). The reflection coefcient
acoustic impedance existing across the interface. (R = Ar/Ai) and transmission coefcient
The fundamental theoretical principles upon (T = At/Ai) for the case of normal incidence, are
which solution of Zoeppritz equations is based given in terms of the contrast in the acoustic
on, is the fulllment of the conditions which impedance (Z) of the two media on either side of
require that all normal and tangential stresses and the interface. Thus, by denition, the reflection
displacements at the interface are continuous. For and transmission coefcients (R and T) are:
displacement continuity, the sum of normal and
tangential displacement-components (at the R Ar =Ai q2 v2  q1 v1 =q2 v2 q1 v1
interface) in the rst medium must be equal to Z2 Z1 =Z2 Z1
the corresponding sum in the second medium. T 1  R At =Ai C=A 2q1 v1 =q2 v2 q1 v1
Concerning stress, it is required that, the sum of 2Z1 =Z2 Z1
normal and tangential stress-components are
similarly equal. Another important note in this
If (R) is positive, compression displacement is
contest is that Zoeppritz equations are based on
reflected as compression because, by reflection in
the assumption that all the involved waves are
this case, both of the wave-motion direction and
pure sine waves of same frequency which drops
particle displacement are reversed. If, however,
out of the equations (Richter 1958, p. 670). This
(R) is negative, a compression is reflected as
means that no frequency change taking place in
dilatation, meaning that we get (180) phase
reflection and transmission processes.
change in this case.

4.2.2 Reflection Coefficient


at Normal Incidence
Pi Pr
The mathematical derivation of Zoeppritz equa-
tions for the case of inclined incidence is of
complicated nature. However, computations
become less complicated in the case of normal
Z1 = 1V1 interface
incidence. A P-wave, for example, hitting an
interface in the direction of the normal to the Z2 = 2V2
plane of the interface (angle of incidence equals
zero) will give rise to only reflected and trans-
mitted P-waves. No wave conversion and no
refraction shall take place in this case (Fig. 4.4).
The efciency of an interface in reflecting Pt
seismic energy is expressed by the reflection
Fig. 4.4 Reflection in case of normal incidence of
coefcient which is dened by the ratio of reflected P-wave at an interface. The incident, reflected, and
amplitude (Ar) to the incident amplitude (Ai). transmitted waves are Pi, Pr and Pt respectively
74 4 Seismic Wave Reflection and Diffraction

It is evident from these equations (case of case where the acoustic impedance in one of the
normal incidence) that the reflection coefcient two adjacent media is approaching zero or innity.
depends only on the contrast in the acoustic Thus, when (Z1) is very small compared with (Z2),
impedance (Z2 Z1). The greater the contrast, that is when (Z1) approaches zero, (R) approaches
the larger reflection coefcient will be. This unity (R = 1). In this case, the interface is consid-
implies that no reflection occurs from an inter- ered to be ideal reflector since this means that all of
face across which the acoustic impedance the incident energy is reflected and no part of it is
assumes the same value, even if either velocity or transmitted through. A good approximation of this
density varies individually across the interface. kind of situation is the earth free-surface which is an
The reflection coefcient ranges in value from interface between the upper air-medium and the
(1) to (+1) depending on the acoustic impe- lower rock-medium. Because of the vast difference
dances (Z1 and Z2). When the acoustic impe- between the acoustic impedances of these two
dance (Z2) in the second medium (medium in media, the reflection coefcient approaches the
which transmission occurs) is greater than that in value of (+1) for a source located in the air and (1)
the rst medium (medium of incident wave, Z1) a for a source located inside the rock-medium. The
compression displacement is reflected as com- interface in both cases is equally efcient in
pression, and the reflection coefcient is positive. the reflecting process, but in the second case there
In the opposite case, that is when (Z1 > Z2), a occurs a polarity reversal, or phase change of (p) for
compression is reflected as rarefaction (phase an incident sine wave.
change = p) and the reflection coefcient
becomes negative. Reflection coefcient is zero
when the two acoustic impedances (Z1 and Z2) 4.2.3 Geometry of Reflection
are equal, which means that there is no interface from Dipping Reflectors
existing in the way of the incident wave.
Another useful feature which may be deduced The general case of reflection geometry is the
from the reflection coefcient expression, is the case of inclined (dipping) reflector (Fig. 4.5).

Fig. 4.5 Ray-path (a) and (a) surface


R S R
travel-time curve (b) of a
seismic wave generated at
the source point (S),
reflected from a dipping
interface, and received by h reflection ray-path
receivers (R). Interface is V1
separating media of
velocities V1 and V2 V2 dipping reflector

(b) x 0 x
Xm

travel-time
apex
curve

T(x)
4.2 Wave Changes at Reflection Interface 75

The travel-time function of a seismic wave Tx x2 =V2 4z2 =V2 1=2


reflected from an interface which is dipping by angle
(h) in the source-to-receiver direction is given by: This is also a hyperbola having its apex
located at the point (0, 2z/V) which is symmet-
Tx x2 =V2 4z2 =V2 4xz sin h=V2 1=2 rical about the time-axis. Using (z = VT0/2), this
becomes:
or,
 1=2
 1=2 Tx x2 =V2 T20
Tx x2 =V2 T20 2x T0 sin h=V
As in the previous case, this equation can be
where (T0 = 2z/V), represents the two-way expressed in an approximate form by using the
reflection travel-time for the case of zero-offset binomial expansion. The reflection function in
receiver (receiver placed at x = 0), and (z) is the this case becomes:
length of the travel-path which is perpendicular
to the dipping reflector. Tx  x2 =2T0 V2 T0
This equation can be expressed in an
approximate form, using the binomial expansion Ray path and the corresponding travel-time
(with truncation of the resulting series after the curve of reflection from a horizontal interface, is
rst term) which gives: shown in the following Fig. 4.6.

Tx  x2 =2T0 V2 T0 x sin h=V


4.2.5 Reflection from Multiple
where (Tx) is the travel-time, (V) the wave Reflectors
velocity, (x) source-receiver distance. The dip
angle (h) is given a minus sign when the dip is in It is possible that a seismic wave arrives at a
the receiver-to-source direction. point on the surface after being reflected several
This is a hyperbola having its apex located at times from a number of interfaces. The rst
the point (xm, Tm) where: arrival (called the primary reflection) is the
strongest, followed by other reflection arrivals
xm 2z sin h; Tm 2z cos h=V (the multiple reflections) which are of less
energy.
This means that the apex of the reflection hyper- A well known type of multiple is the (ghost
bola is shifted by (2z sin h) in the up-dip direction for reflection) which occurs when a wave travels
a dipping reflector with angle of dip (h). It should be upwards from a source-point located at a certain
noted here that (h) denotes the apparent dipping in the depth and is reflected by the free earth surface or
source-receiver direction. Determination of the true by the base of a low-velocity surface layer (the
dip (i.e. maximum dip) can be achieved using two weathering zone). Because of the short extra
apparent dips measured in two different directions travel path, the multiple arrives at a short
(see for example Nettleton 1940). time-interval after the primary reflection. On the
seismic section it appears as a weak reflection
event which is closely following the stronger
4.2.4 Geometry of Reflection primary reflection, and for this appearance it was
from Horizontal named ghost reflection.
Reflectors Several other types of multiples may occur
depending on the structural form of the medium.
For a horizontal reflector the travel-time function Multiple reflections include also reflected
is obtained by putting (h = 0) in the function for refraction or refracted reflection. Identication of
the dipping reflector, giving: these wave arrivals is very important, since
76 4 Seismic Wave Reflection and Diffraction

Fig. 4.6 Ray-path and surface


travel-time curve of a R S R
seismic wave reflected
from a horizontal interface.
S and R are source and reflection
receiver points respectively ray-path
h

V1 horizontal reflector
V2

x 0 x

apex
travel-time
curve

T(x)

mistaking a multiple for a primary reflection DT Tx T0


introduces a serious error in the interpretation
results. Some of the commonly known types of We shall now consider the cases which are
multiples are shown in Fig. 4.7. involved in NMO computation for a single hori-
zontal reflector and multiple horizontal reflectors:
4.3 The NMO and DMO Concepts (i) NMO in Case of Horizontal Reflector
For a horizontal reflector, the NMO equation
The reflection travel-time function plays a funda- takes the following form:
mental role in seismic exploration. With velocity
h i1=2
data, the travel time function can give valuable
DT Tx T0 x=V2 T20 T0
information of the subsurface geological structure.
Closely associated with the reflection travel time
function are other similar functions which have This equation can be expressed in another
equally important applications in the seismic form, using the binomial expansion (with trun-
reflection exploration. These are the Normal cation of the resulting series after the rst term).
Move-Out (NMO) and the Dip Move-Out (DMO). This form gives:

DT x2 =2T0 V2
4.3.1 The Normal Move-Out (NMO)
Concept This is considered as an accepted approxi-
mation since (x/VT0  1), which is usually the
The Normal Move-Out (NMO) is dened as the case in seismic reflection exploration. In this
difference (T) between reflection travel-time mathematical process, the exact form of (T) is
(Tx) and the two-way vertical travel-time (T0). transformed from its hyperbolic function to the
For a single horizontal reflector found at the base approximate form which is a parabolic function.
of a homogenous layer (of constant velocity), the Both forms of these two equations show that
NMO parameter (T) is given by Fig. 4.8. (T) is function of three variables; the receiver
4.3 The NMO and DMO Concepts 77

primary long-path multiples short-path multiples

surface

reflector-1 ghost ghost

reflector-2

reflector-3

Fig. 4.7 Types of multiple reflections

(a) (b) (c)


S/R S R x
x 0
T0

Tx
T
z

reflector
Fig. 4.8 The NMO concept. a Two-way vertical ray-path. b Two-way slant ray-path. c Reflection arrival times
corresponding to the ray-paths of a and b. The NMO (DT) is shown as the difference, (Tx T0)

T T T

x T0 V

Fig. 4.9 NMO (DT-function) direct variation with (x) and inverse variation with both T0 and V

offset (x), velocity (V), and the two-way vertical For a given offset (x), the time (Tx) can be
time (T0). The proportionality is direct with (x2) readily measured for a certain reflection event
and inverse with both (V) and (T0). These appearing on the seismic trace recorded at that
changes are shown as follows Fig. 4.9. offset. From these data, the NMO (T) can be
78 4 Seismic Wave Reflection and Diffraction

accurately calculated (from the given NMO equa- h i1=2


tions) provided that the velocity (V) is known. Tx x=Vr 2 T20
Application of the NMO concept is centered DT x2 =2T0 V2r
on velocity determination which is normally
determined by a special processing step, the ve- where (Vr) is the root mean square velocity for
locity analysis process (Sect. 10.7.1). the whole traversed multi-layer medium. The
(ii) NMO in Case of Multiple Horizontal velocity (Vr) can be computed from the interval
Reflectors velocities of individual layers using the formula
In an n-layer medium, the ray reflected from the given in Chap. 3, Sect. (3.4.4).
base of the nth layer will arrive at the receiver (iii) NMO in Case of a Dipping Reflector
placed on the surface, after it has been refracted The travel-time of a wave reflected from a given
at each interface met with during the total horizontal reflector is least in the case where the
reflection travel-path (Fig. 4.10). source and receiver occupy a common position,
The difference between this case and the sin- that is in the zero-offset (x = 0) situation. Thus,
gle layer case is that here we have more than one as the offset increases the travel time increase. In
velocity with which the reflection wave has other words, the NMO increases with the offset
travelled. The NMO-velocity, in this case relative to the vertical reflection time. When the
(multi-layer case) must be a sort of average or reflector is dipping the increase in travel-time
effective velocity depending on the individual becomes function of both the offset (x) and dip
velocities of the multi-layer medium. According (h). This is expressed by the dipping-reflector
to (Dix 1955), the root mean square velocity travel-time equation. Thus:
should be used in this case as its value is the
closest to the effective velocity for the multi-layer  1=2
Tx x2 =V2 T20 2x T0 sin h=V ; exact form
medium. The RMS velocity can be derived from
the interval velocities of the layers which are or,
making up the geological section existing above
the nth reflector in this case. Thus, the functions Tx  x2 =2T0 V2 T0 x sin h=V; approximate form
for the reflection travel-time (Tx) and NMO
(T) for multi-reflector case will be in the form:

Fig. 4.10 Ray-path x


geometry of reflected 0
waves, in a three-reflector surface
model. Within each layer
the ray-path is straight line
because velocity is
assumed to be constant for V1
the layer reflector-1
Tx T1
V2 z
reflector-2
T2
V3
reflector-3
T3
V4
4.3 The NMO and DMO Concepts 79

Fig. 4.11 Reflection (a) (b)


ray-paths and travel-time 0 +X -X
-X 0 +X
curves for two cases;
apex
a dipping reflector and
b horizontal reflector. S and T apex travel-time
R are source and receiver curve
points respectively

Tx Tx

R S R R S R
surface

reflection
ray-path

In the case of a horizontal reflector, reflection DTd 2x sin h=V


travel-times received at two receivers of equal
offsets are equal, whereas, in the case of a dip- Or, for small h (in radians) we can write,
ping reflector, the travel-times of rays received at
DTd 2x h=V
two receivers of equal offsets are not equal. In
this case (case of two receivers of equal offsets),
This can be used in calculating the dip angle
and because of dip, the ray-path in the down-dip
(h), where (h = VTd/2x).
direction is longer than that in the up-dip direc-
tion (Fig. 4.11).
4.4 The CDP, CRP, and CMP
Concepts
4.3.2 The Dip Move-Out (DMO)
Concept
According to the simple laws of reflection, the
reflection point is located vertically below the
Dip move-out (Td) is dened as the difference
source-receiver mid-point. This point is given
in travel-time (T+x) and (Tx) of rays reflected
more than one name, depending on its identity
from a dipping reflector to receivers of equal and
criteria. It is called reflection point (RP) because
opposite offsets; (+x and x) (Kearey and
it is the point where reflection occurs, and depth
Brooks 2002, p. 46). That is:
point (DP) or mid-point (MP) because its posi-
DTd T x  Tx tion is at a subsurface interface or because its
position is vertically below the source-to-receiver
Using the reflection travel-time equation (the mid-point. All these three points coincide when
approximate form), we get: reflector is horizontal (Fig. 4.12).
80 4 Seismic Wave Reflection and Diffraction

S R surface S
S3 S2 S1 R R1 R2 R3

reflection ray-path
horizontal reflector
CDP
horizontal reflector CRP
DP CMP
RP
MP Fig. 4.13 Denition of the common points (CDP),
(CRP), and (CMP), which coincide on each other in case
Fig. 4.12 Denition of the depth point (DP), located of horizontal reflector
vertically below the mid-point of the source-receiver
distance. It is the reflection point (RP), or mid-point (MP)
surveying. Application of the concept usually
results in great enhancement of the reflection
These points are described as common points signal by summing together (stacking) the
(CDP, CRP, CMP) in cases where multi- CDP-Gather traces.
reflections occur for the particular point. In other
words, a point is described as common when it
becomes common to more than one reflection. 4.4.2 CDP, CRP, and CMP in Case
of Dipping Reflector

4.4.1 CDP, CRP, and CMP in Case For a horizontal plane reflector, the points (DP,
of Horizontal Reflector RP and MP) coincide at a point which is located
vertically below source-to-receiver midpoint. In
By use of certain source-receiver layout, it is the case of dipping reflector, however, the
possible to shoot a number of shots such that the reflection point gets shifted up-dip by a distance
reflection points resulting from this number of depending on the offset as well as on the dip
shots coincide on each other, that is one angle of the reflector. The up-dip shift is given
point-location will serve as common point to all by:
of the implemented shots. In this type of repeated
shooting-spread, the reflection point becomes DL x2 cos h sin h=D
known as common depth point (CDP), common
reflection point (CRP), or common mid-point where (L) is the up-dip shift of the reflection
(CMP). These points are shown in Fig. 4.13. point (RP), (x) is half the source-receiver offset,
The four reflection ray-paths, shown in (h) is reflector dip-angle, and (D) is the length of
Fig. 4.13, have one common depth point (the reflection ray-path which is normal to the
CDP), will each produce a seismic trace. These reflector (Deregowski 1986). The ray-path
traces belong to the same depth point (CDP) and geometry and location of the reflection point
thus they form a group of traces called the (RP) are shown in Fig. 4.14.
(CDP-Gather). In this case (case of horizontal In case of repeated shooting (as it is done in
reflector) the CDP and the other points (CRP and the usual seismic reflection proling surveying),
CMP) will coincide on each other. the reflection points (RPs) get dispersed
Ever since the CDP concept was introduced along the dipping reflector-plane. For a given
by Mayne (1962, 1967) it has been applied on a reflector, the amount of up-dip shift (L)
routine basis in execution of seismic reflection increases with the square of the receiver-offset
4.4 The CDP, CRP, and CMP Concepts 81

S MP R in pre-stack migration), reflection arrivals of a


surface CMP-gather traces, from a dipping reflector, are
corrected such that the end result will be corre-
sponding to rays being reflected from a common
Reflection reflection point (CRP) located at the CMP of that
D
ray-path dipping
gather. In other words, the dispersal of the RPs of
non-zero offset ray-paths is removed (Fig. 4.16).
RP
reflector The CMP concept is fundamental issue of the
L seismic reflection proling technique. This is
because it covers the general case of the geom-
etry of the reflection interfaces, regardless of
Fig. 4.14 Up-dip shifting of reflection point (RP) in case
of dipping reflector. S and R are source and receiver points
their dips.
respectively and DL is the up-dip shift of the RP An important step carried out in seismic data
processing is stacking (summing up) of the CMP
seismic traces (normally called the CMP-gather).
and inversely proportional to the normal
In case of dip, the CDP and CRP are no longer
midpoint-reflection distance (D). Spreading out
applicable, whereas, the CMP principle is appli-
of the reflection points in the up-dip direction is
cable to all dipping and horizontal reflectors as
shown in Fig. 4.15.
well. The role of the CMP concept shall be more
Referring to Fig. 4.15, the four reflection
claried in the coming chapters.
ray-paths, having one common point (the CMP),
will each produce a seismic trace. These traces
are reflected from dispersed reflection points
(RP1, RP2, RP3, and RP) and not from one point 4.5 The Seismic Wave Diffraction
as in the case of a horizontal reflector. For this
type of source-receiver set-up, there is one point A part of a seismic wave-energy is reflected
that is common to all of these ray-paths, which is when the wave hits a continuous plane interface
the source-receiver midpoint (the CMP). Thus, surface. However, when the interface is not a
the group of traces which belongs to common continuous plane but of curvature which is large
point is the group of traces belonging to the in comparison with the curvature of the incident
common mid-point (the CMP-Gather). wave-front, the change in propagation direction
An important note is worth mentioning here, it does not follow the known laws of reflection. If
is that, with this type of source-receiver layout, the intervening obstacle, to an advancing seismic
we cannot have a common reflection point wave, is of size approaching to a point-
(CRP). However, in processing (as in DMO, and reflector, the wave radiates from that obstacle,

Fig. 4.15 Denition of S


the common midpoint R
(CMP) obtained from
S3 S2 S1 R1 R2 R3
multi-reflection CMP
experiment, in case of a
dipping reflector D

RP3
RP2
RP RP1


82 4 Seismic Wave Reflection and Diffraction

S
S3 S2 S1 R R1 R2 R3
CMP

CRP


Fig. 4.16 Denition of the common reflection point (CRP) obtained from multi-reflection experiment. For dipping
reflector, the (CRP) is located at one reflection point which is coincident with the end point of the normal from the CMP

Fig. 4.17 Generation of a


diffracted wave with
spherical wave fronts as an
advancing plane seismic
wave hits a small obstacle
(point-diffractor)
advancing point-
plane seismic diffractor
wave

diffracted wave
spherical wave fronts

in every possible direction giving a phenomenon this type is normally referred to as a point-
called (diffraction) and the wave which leaves the diffractor.
obstacle after incidence is called (diffracted The diffraction phenomenon can be explained
wave). by considering the diffraction point as a
A closely related term is (wave scattering) point-source which, upon being hit by an incident
which is used to describe diffracted wave-eld wave, becomes an activated source that radiates
caused by small structural irregularities, as for waves in all directions. According to Huygens
example found with seismic energy reflected principle, an obstacle hit by an incident wave
from rugged basement surfaces. becomes an energy-source from which seismic
waves are generated and transmitted in all possible
directions. If a plane seismic wave, for instance hits
4.5.1 The Point-Diffractor a point-diffractor embedded in a homogeneous
medium, a diffracted wave with spherical wave
A structural obstacle, having radius of curva- fronts will be generated. The diffracted wave will
ture which is shorter than that of the incident move away from that source-point causing inter-
wavelength, acts as a diffraction-generating ference with the incident wave-train and all other
point (or diffraction point). An obstacle of waves that may be coexisting at the time (Fig. 4.17).
4.5 The Seismic Wave Diffraction 83

4.5.2 Diffraction from Terminating diffraction hyperbola is opposite to that in the


Reflectors backward branch. As regards the waveform
(wave spectrum structure) of the diffracted wave,
One of the most common examples of diffraction there is a general decrease in high-frequency
which are met with in seismic exploration, are those content which, like amplitude behavior, gets
created from discontinuous reflectors as they occur more severely attenuated as the travel distance
when formations are faulted. Thus, when a plane increases (McQuillin et al. 1984, p. 24).
seismic wave is normally incident on the surface of In case of diffraction from the termination edge
a faulted geological bed, it will be reflected from of a reflector, the polarity of the diffracted wave is
that surface and diffracted from the reflector reversed (180 phase change) in the forward
termination-end which is acting as a diffraction branch of the diffraction hyperbola compared to
source or diffraction-generating point (Fig. 4.18). that in the backward branch. The terms, forward
Reflector termination, as in the case of a fault and backward, branches of the hyperbola are used
or a pinchout, for example, acts like a point to specify the two branches of the diffraction
diffractor, whose response, on a zero-offset traces hyperbola. The forward branch is the branch
(such as a stack section), is a diffraction hyper- which tends to carry the reflection forward, and the
bola, which is also called the curve of maximum backward branch is the branch that lies underneath
convexity. the reflection image (Telford et al. 1990, p. 178).
Diffraction ray-path from a terminating
diffracting edge and the corresponding seismic
4.5.3 Diffraction Seismic Image image (diffraction hyperbola) is schematically
illustrated in Fig. 4.19.
The amplitude of diffraction wave attains its Discrimination of diffraction events on seismic
maximum value at the detector located vertically records (diffraction seismic image), based on
above the termination point of the reflector, and it waveform characteristics, is practically not possi-
decreases with increasing offset-distance from ble. However, identication of diffraction is pos-
this point. The rate of attenuation of a diffracted sible by its characteristic travel-time curve.
wave with increasing detector-offset is greater Diffraction travel-time curve is different from those
than that of normal reflection wave in the same of the other types of waves as it will be explained in
medium. We may also note that the polarity of the following discussion concerning the travel-path
the diffraction arrival in the forward branch of the geometry of the propagating diffraction waves.

Fig. 4.18 Occurrence of


diffracted wave due to a advancing plane seismic wave
normal-incident plane wave
onto a terminating reflector.
Wave fronts and rays of
these waves are drawn

reflected wave

incident wave

terminating reflector
terminating reflector
fl t point-
point-diffractor
Point
point diffractor
diffractor
diffracted
spherical
wave front
84 4 Seismic Wave Reflection and Diffraction

Fig. 4.19 Sketch (a)


showing, a terminating incident wave diffracted wave
reflector model with
vertical incidence and
diffraction ray-paths and
b its seismic image for both reflected wave
reflection and diffraction
arrivals point-diffractor
plane reflector

(b) reflection arrivals

-ve diffraction arrivals +ve

backward branch forward branch

Fig. 4.20 Reflection and x


diffraction ray-paths, with
S R S R
x
their respective travel-time
curves. The diffraction
normal move-out (DTd) is reflection
greater than the reflection diffraction
ray-path
normal move-out (DTr) for
the same offset (x)
ray-path
i r
terminating reflector
diffraction point
reflection points

0 0

Tr Tr Td Td

reflection travel-time diffraction travel-time

4.5.4 Diffraction Travel-Time normal move-out. To compare diffraction hyper-


Function bola with the reflection hyperbola for a given in-
terface, consider the terminating reflector shown
Identication of diffraction is possible by use of its in Fig. 4.20.
travel-time curve, which, like reflection From the geometry of the reflection ray-path
travel-time, is of hyperbolic form, but with larger and for diffraction ray-path from its termination
4.5 The Seismic Wave Diffraction 85

Fig. 4.21 Point-source diffraction arrivals x


diffraction ray-path and its
seismic image, which is a
hyperbolic curve centered ray-path
at the point diffractor

depth
point source
x

T0
Tx
seismic image

diffraction
time hyperbola

edge, we have for the same offset (x), the travel and,
time function (TDx) for diffraction, and (TRx) for
the reflection, are given by: DTd  x2 =T0 v2
h i1=2 giving,
TDx x=v2 T0 =22
h i1=2 DTd  2DTr
TRx x=v2 T20
This shows that the normal move-out of a
and, wave diffracted from a terminating reflector is
approximately equal to double that of a wave
h i1=2
DTd TDx T0 x=v2 T0 =22 T0 =2 reflected from the same interface at the same
h i1=2 offset. This important feature is used in dis-
DTr TRx T0 x=v2 T20 T0 criminating diffraction from reflection events.

where, (Tr) and (Td) are the normal move-out 4.5.5 The Diffraction Hyperbola
for the reflection and diffraction respectively, and
(v) is the propagation velocity. For small The seismic image of the diffraction arrivals is (for
offset-to-depth ratio, that is for (x/vT0 < 1), these constant velocity) a hyperbolic curve centered
equations can be approximated by: about the diffraction source-point. A diffraction
event such as this is expected to appear on a seis-
DTr  x2 =2T0 v2 mic section made up of zero-offset traces. The
86 4 Seismic Wave Reflection and Diffraction

Fig. 4.22 The seismic depth model time image


response of a diffraction
point is a hyperbolic curve,
the diffraction hyperbola
surface surface

diffraction point
diffraction hyperbola

seismic stack section is effectively a section of of their interferences, these arrivals are distorting
zero-offset traces (Fig. 4.21). or masking the reflection events and causing
Diffraction hyperbolae are often observed on smearing effects at the fault-zone leading to
seismic stack sections, indicating point-source decrease in fault-resolution. Example of the
diffractions, as those generated by faults and interferences introduced as a result of fault-
pinchouts. A seismic diffraction event (diffrac- generated diffraction hyperbolae are shown in
tion hyperbola) appearing on seismic stack sec- Fig. 4.23.
tion can be considered as the seismic response of Although diffraction events (diffraction
a depth section of a homogeneous medium con- hyperbolae) are introducing masking effects to
taining a point-source diffractor (Fig. 4.22). the primarily targeted reflection events, they
(diffraction events) can help interpreters in
identifying faults and other diffraction-sources.
4.5.6 Distortion Effects of Diffraction In general, diffraction events on stack sections
are considered to be unwanted distortive events
In nature interfaces are not always plane and which must be removed or at least attenuated as
continuous surfaces. There are cases where these much as possible. The common way to remove
surfaces are irregular as the surfaces of reef the diffraction events (diffraction hyperbolae) is
bodies or discontinuous as the faulted beds or by application of the seismic migration which is
pinchouts as found with angular unconformities. one of the principal steps applied in processing of
When a seismic wave is incident on such sub- seismic reflection data.
surface features, diffracted waves are generated
and transmitted through the medium, interfering
with, and distorting other co-existing waves such 4.5.7 The Exploding-Reflector Model
as reflection events, the main objective in normal
seismic exploration. In accordance to Huygens principle, the
On a seismic stack section, which is effec- exploding reflector model is a model in which a
tively made up of zero-offset traces, the diffrac- reflection interface is considered to be formed of
tion arrivals (diffraction event) appear as a innite number of closely packed diffraction
hyperbolic curve whose apex is coincident with points and that these points explode at a given
the causing diffraction point. Faulting is one of start-time, generating seismic waves. The
the principal sources of diffraction waves seen in exploding reflector modeling process requires
stack sections. A reflector termination caused by that each point of the refelector is sending seis-
faulting, generates diffraction arrivals. Because mic rays, in every direction (exploding) at a
4.5 The Seismic Wave Diffraction 87

Fig. 4.23 Distortions due


to diffraction hyperbolae,
created by a faulted ray-path
reflector, from two
diffraction points. The
terminating reflector
forward and backward
branches show polarity
reversal
terminating reflector

seismic

image

diffraction hyperbolae

forward branch backward branch

Fig. 4.24 The exploding receiver


reflector model. Seismic surface
energy transmitted from the
diffracion points
simultaneously and
recorded by receivers
normal
placed on the surface diffracted
seismic rays
seismic
rays
reflector diffraction points

common start time. The moving seismic ampli- wave fronts of waves diffracted from closely
tude is considered to be of magnitude propor- packed diffraction points (making up the reflec-
tional to the normal-incidence reflection tion surface), is called exploding reflector model.
coefcient. The resulting wave eld generated according
The generated waves are made to propagate to to the exploding-reflector model shall propagate
the surface with velocity half that of the actual along the normal-incidence ray-paths. The con-
velocity. With this velocity-value the one-way cept is schematically shown in Fig. 4.24.
travel-time to the surface becomes equal to the Compared with the conventional record sec-
two-way reflection travel-time for zero-offset tion, the section produced from the exploding
receivers placed on the surface (Sheriff 2002, reflector model, is the same as the conventional
p. 127). Consideration of the reflection process as section having propagation velocity equal to half
being formed from the interactions of seismic of the true velocity (Fig. 4.25).
88 4 Seismic Wave Reflection and Diffraction

Fig. 4.25 Sketch showing (b)


(a)
two cases of zero-offset exploding
conventional
trace and its corresponding S reflector S
reflector
rapath. a Conventional R model R
model
two-way travel-time,
b exploding reflector two-way one-way
model, one-way travel-time time time
with model velocity equal
to half that of the true vmodel = v vmodel = v/2
velocity

reflection point diffraction point

Application of the exploding reflector concept in of migration (both of the pre- and post-stack
computing the seismic reflection section is a migration processing) using the exploding-reflector
direct-modeling (forward-modeling) process. concept. This approach proved to result in great
Algorithms have been developed for computations reduction of the migration computational cost.
Seismic Wave Transmission
and Refraction 5

The term (transmission) is customarily used to layers down to a reflector and up to the surface.
indicate the general case of wave propagation Among other factors, reflection and transmission
where the moving seismic wave crosses an coefcients cause attenuation to the wave energy
interface whether the incidence is normal or as it is experienced by the recorded function r(t).
inclined. When an obliquely incident wave is It should be remembered that there is a reflection
transmitted across an interface, it is bent at the and transmission processes taking place at each
interface towards, or away from, the normal at interface existing in the traversed geological
the point of incidence. This is the well known section (Fig. 5.1).
phenomenon of (refraction). A refracted wave is,
therefore a transmitted wave resulting from an
inclined incidence. No refraction occurs when 5.1.1 Transmission Coefficient
the wave is incident perpendicularly at an inter- at Normal Incidence
face. Thus, it can be sayed that the refraction
phenomenon is a special case of transmission. It The transmission coefcient (or transmittance,
(refraction) occurs only with normal incidence. as it is sometimes called) is dened to be the
ratio of the amplitude of the transmitted wave to
that of the incident wave. Sometimes this is
5.1 Seismic Wave Transmission expressed in terms of energy instead of ampli-
tude. At normal incidence, the transmission
When a seismic wave hits an interface, part of the coefcient (T), expressed as a ratio of the
wave energy is reflected and the rest is transmitted transmitted amplitude (At) to the incident
across that interface. The amount of energy which amplitude (Ai), thus:
is reflected from an interface is determined by the
reflection coefcient and what is left from the T At =Ai
incident energy is transmitted across the interface
into the second medium. At the interface the As it is with the reflection coefcient, the
ray-path is bent if the incidence is oblique and transmission coefcient is function of the
continues with unchanged direction when the acoustic impedances of the two media separated
incidence is normal to the interface. by the involved interface. For an interface sepa-
In the real layered Earth crust, the initiated rating two media of acoustic impedances (Z1 and
seismic energy, the source function s(t), is Z2), the transmission coefcient (T), is related to
recorded as r(t) after it penetrates all the rock the impedances by:

Springer International Publishing Switzerland 2017 89


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_5
90 5 Seismic Wave Transmission and Refraction

s(t) r(t) T 1R

t t For a given interface, the (T-R) relationship is


0 0
(T1 = 1 R) for a wave incident (normal inci-
dent) from the low-impedance (Z1medium) to
surface
the high-impedance (Z2medium), and
(T2 = 1 + R) for transmission in opposite
R1 direction. On this basis the two-way transmission
coefcient (call it T12) for an interface penetrated
R2 twice (one is in opposite direction to the other) is
R3 given by the product (T1  T2), that is:

R4 T12 1  R  1 R 1  R2

Fig. 5.1 Wave-attenuation due to reflection and trans- And in terms of impedances it is given by:
mission processes which are normally occurring in a real
layered rock medium. s(t) and r(t) are source receiver T12 4 Z1 Z2 =Z2 Z1 2
time-functions respectively
The two-way transmission coefcient (T12) is
useful in computing the effective attenuation
T 2Z1 =Z2 Z1
factor of a wave reflected from a subsurface in-
It is apparent from this formula that as the terface after being transmitted through several
contrast in the acoustic impedances is smaller the layers during its total reflection travel path.
transmission coefcient becomes greater, and in
the limit when they are equal (Z1 = Z2), trans- 5.1.3 Attenuation Due to Reflection
mission coefcient (T) becomes unity and all of and Transmission
the seismic energy shall pass through and no
reflected energy will take place. For an n-layer medium, the wave which is
For the case of inclined incidence, the trans- reflected from the nth interface and received at
mission coefcient, like the reflection coefcient, the surface would have crossed the (n 1)
depends on both of the angle of incidence and on interfaces twice. Assuming normal incidence, the
the contrast in the acoustic impedance of the effective attenuation factor (RTn) of a wave
adjacent media. Unlike reflection, however, there reflected from the base of the nth layer (nth
is no special situation whereby phase change subsurface interface) and transmitted through
occurs. (n 1) interfaces will be given by:
RTn An =Ai
5.1.2 The Two-Way Transmission
Coefficient By use of the concept of the two-way trans-
mission coefcient, it is possible to derive a for-
The transmission coefcient (T) is related to the mula that computes the net attenuation due to the
reflection coefcient (R) by the expression (see combined effects of the reflection and transmis-
Sect. 4.2.2): sion coefcients. In a multi-layer geological model
5.1 Seismic Wave Transmission 91

Fig. 5.2 Reflection and


transmission ray-path of a Ai A1 A2 A3 A4 = Ai R4 (1-R21) (1-R22) (1-R23)
wave traversing a 4-layer
medium. (A4) is
wave-amplitude reflected
from the 4th reflector
(vertical incident is R1
assumed)
R2

R3

R4

(n layers), the reflection-transmission process of function of the reflection coefcients of the


an initial incident amplitude (Ai) will experience existing n-reflectors (Rn). For vertical incidence,
attenuation due to the double crossings which it is given by:
occur at (n 1) interfaces. In a multi-layer model,      
the amplitude (An) is received at the surface after RTn Rn 1  R21 1  R22 1  R23 . . . 1  R2n1
covering the whole of the reflection travel path,
from the point of the incident amplitude (Ai) to the With knowledge of the number of reflectors in
receiver point. Ray-path geometry of such a a section and their reflection coefcients, it is
multi-layer model is shown in Fig. 5.2. possible to compute the extent of attenuation a
Applying the two-way transmission coef- reflected wave experiences in covering the com-
cient we can write: plete source-to-receiver reflection-travel-path.

A1 Ai R1
  5.1.4 Role of Transmission in Seismic
A2 Ai R2 1  R21 Exploration
  
A3 Ai R3 1  R21 1  R22
... Seismic wave transmission surveys make use of
      seismic waves which are directly propagating
An Ai Rn 1  R21 1  R22 1  R23 . . . 1  R2n1 from sources to receivers. The normal direct
wave, which is always recognized on seismic
A general recursive formula can be derived by reflection and refraction-shooting records, is a
dividing the amplitude of the nth reflector by that typical transmission method used to calculate the
of the (n 1) reflector, giving: velocity of the surface layer. In exploration
  seismology, there are several transmission-based
An An1 Rn Rn =Rn1 1  R2n1 ; n2 methods which are applied for specic explo-
ration purposes. Up-hole surveying, well velocity
Repeating, this formula is valid for (n  2), surveying, vertical seismic proling (VSP) fan
and for (n = 1), we have A1 = Ai R1. shooting, and seismic transmission tomography,
In conclusion the effective attenuation factor are typical examples of exploration practices
RTn (=An/Ai) due to reflection-transmission employing the seismic wave transmission phe-
process for a wave reflected from the nth nomenon. The up-hole surveying and that of
reflector of a multi-layer geological section, is well-velocity are dened here-below (Fig. 5.3).
92 5 Seismic Wave Transmission and Refraction

Fig. 5.3 Examples of (a) (b)


exploration activities drill-hole drill-hole
employing seismic wave
seismic
transmission. a An up-hole detector
source
survey and b well velocity
survey

seismic
source detector

(i) Up-hole Surveying amount of bending (angle of refraction) and the


In principle, the eld arrangement is similar to sense of bending (towards or away from the
that used in the well velocity surveying. The normal) are governed by the velocities of the two
difference is that, in the up-hole surveying the layers separated by the interface. The refracted
source points are placed at different depths inside ray is bent away from the normal in case of the
the drill-hole, while the detector is placed on the velocity in the medium in which the wave is
surface, close to the head of the bore-hole. refracted) is greater than that in the medium that
hosts the incident wave. The bending of the
(ii) Well Velocity Surveying
refracted wave is towards the normal when the
The normal survey procedure starts with placing
velocity of the medium, in which refraction
the detector at a certain depth inside the well,
occurs, is of lower value (Fig. 5.4).
while the shot is on the surface. The
In essence, the phenomenon of ray bending
source-to-detector direct-arrival is recorded.
(wave refraction) at an interface occurs only
From the transmission travel times, corrected to
when the incidence direction is inclined to the
the vertical path, the average velocities are cal-
interface. No bending ocurrs with perpendicular
culated and plotted against depth.
incidence.
It should be noted here that the transmitted
waves are, in general, experiencing refraction of
various severities, since, in nature, no perfectly
normal incidence to interfaces are expected to 5.2.1 Snells Law of Refraction
occur during the wave propagation.
Where there is a change in the propagation
direction on crossing the interface, the transmit-
5.2 Seismic Wave Refraction ted wave is normally referred to as refracted
wave. This takes place only when the incidence
A seismic wave incident on an interface is partly is inclined with respect to the interface plane.
reflected and partly transmitted. In the general The degree of the ray bending (expressed by the
case where the wave is obliquely incident on the angle of refraction) depends on the velocity
interface, the transmitted wave changes direction contrast of the media on either side of the inter-
(refracted). It will be bent either towards the face. Refraction is an expression for geometrical
normal line (normal to the interface at point of changes of the wave ray-path on crossing an
incidence) or away from that normal. The interface and not for changes in energy content.
5.2 Seismic Wave Refraction 93

(a) (b) (c)


normal incidence Inclined incidence
no refraction with refraction

i r
M1 (V1 ) M1 (V1 ) M1 (V1 )
M2 (V2 ) M2 (V2 ) M2 (V2 )
rr i

i<r i>r
i=r=0

Fig. 5.4 Occurrence of refraction in case of oblique direction reversed. The velocities of the two media are V1
incidence. a Normal incidence, no refraction, b oblique and V2, where V1 < V2
incidence, from medium M1 to medium M2, c incidence

Fig. 5.5 Reflection and sin i / sin r = v1 / v2


refraction of a seismic
wave at an interface
separating media S R
(M1, velocity v1) and
(M2, velocity v2) where
v2 > v1
incident reflected
wave wave

i i
medium1 velocity v1
medium2 velocity v2
r

refracted
wave

Refraction of an incident seismic wave at an sin i=sin r v1 =v2


interface is governed by Snells law which states
that the ratio of the sine of angle of incidence Referring to Fig. 5.6, the wave front (AB) of a
(i) to the sine of angle of refraction (r) is equal to plane seismic wave is hitting an interface existing
the ratio of velocity in the rst medium (in which between the two media (M1 and M2), at an
the wave is incident) to that of the second med- incidence angle (i). Point (B) requires an interval
ium (in which the wave is refracted). In reference of time (dt, say) to reach the interface after point
to Fig. 5.5, Snells law takes the following form: (A) has reached it. During this interval, the wave
94 5 Seismic Wave Transmission and Refraction

triangles (ABD and ACD), we can write (sin


i = BD/AD) and (sin r = AC/AD), giving the
following relationship:
E B
i i sin i=sin r BD=AC v1  dt=v2  dt v1 =v2
i i
medium, M1 A velocity v1
medium, M2 that is,
r D r velocity v2

C sin i=sin r v1 =v2

which is the well-known Snells law that governs


Fig. 5.6 Refraction geometry of a seismic plane wave at the direction of the refracted wave relative to that
an interface separating two media; M1, velocity v1 and M2, of the incident wave.
velocity v2 The generalized form of Snells law expresses
the relation connecting the reflection and refrac-
front would have travelled the distance of (v1.dt) tion angles with that of the incidence wave, for
into the medium (M1) and (v2  dt) into the all of the wave types resulting from an inclined
second medium (M2). Being a plane wave, the incidence. Thus if a P-wave is obliquely incident
wave front is plane surface which is represented on an interface separating two solid media, four
by the line (AB) for the incident wave, (DE) for wave-types are generated. These are two reflec-
the reflected wave in medium (M1) and line ted waves (P- and SV-wave) and two, transmitted
(CD) for the refracted wave in medium (M2). (refracted) waves (P- and SV-wave), as shown in
From the geometry of the wave path we note Fig. 5.7.
that the two right angle triangles (ABD) and In the above example, if SV is the incident
(AED) are equal, with (AE = BD = v1  dt), wave instead of the P-wave, four similar waves
giving equal reflection and incidence angles in are also generated: two reflected waves (SV and
support of the law of reflection. In the same P) and two refracted waves (SV and P). It is not
Fig. 5.6, by examining the two right-angle always all the four wave-types are generated at

SV
P
P
rsv1

ip1 : P-wave angle of incidence


ip1 rp1 rp1 : P-wave angle of reflection
rsv1 :SV-wave angle of reflection
Zp1 = 1 Vp1
Zsv1 = 1 Vsv1 medium-M1
Zp2 = 2 Vp2 medium-M2
Zsv2 = 2 Vsv2 tsv2
tp2 : P-wave angle of reflection
P
tp2 tsv2 :SV-wave angle of reflection
SV

Fig. 5.7 An obliquely incident P-wave and the generated four wave phases: reflected and refracted P- and SV-waves
5.2 Seismic Wave Refraction 95

an interface. If medium-2, for instance, is liquid, incidence angle (i) increases, and when it reaches
no transmitted SV-wave is formed. a value of (i = ic), refraction angle reaches the
Using the symbols shown in Fig. 5.7, the value (r = 90). The angle of incidence (ic) for
generalized Snells law is quoted as follows which the angle of refraction (r) becomes (90) is
(Sheriff 1973, p. 200):

(sin ip1)/vp1 = (sin rp1)/vp1 = (sin rsv1)/vsv1 = (sin tp2)/vp2 = (sin tsv2)/vsv2 = k
incident P reflected P&SV refracted P&SV

where (k) is constant for the given layer-set, as


the two media (M1 and M2), adopted in this called (the critical angle for refraction). For greater
example. Velocities in (M2) are assumed to be values of this angle there will be no refraction of
higher than those in (M1). the wave into the second medium, and in this case,
We may note that angles of incidence and the incident wave will be totally reflected.
reflection are equal only when the reflected wave is Since (sin 90) is equal to unity, Snells law,
of the same type as that of the incident. Also, no wave at the critical angle (ic), becomes:
conversion occurs when the obliquely incident wave
is SH-wave or when the wave (of any type) which is sin ic v1 =v2
of normal incidence. With the wave conversion
phenomenon in mind, one can envisage the com- Hence, the critical angle of incidence, for any
plexity of the seismic wave-eld developed in an specic interface, will be given by:
inhomogeneous and stratied subsurface medium.
ic sin1 v1 =v2

5.2.2 The Critical Refraction and In this state, where the incident angle is equal
Head Wave Generation to the critical angle (ic), a special refracted wave
(called the head wave) is generated. This criti-
Consider a geological model made up of a med- cally refracted wave travels along the interface
ium (M1) of velocity (v1) overlying a medium with the velocity (v2) of the second medium
(M2) of higher-velocity (v2). An obliquely- (M2). This wave is refracted back to the earths
incident wave in the rst medium shall refract surface at the same angle (ic) and with the
on hitting the separating interface. In such a propagation velocity of the upper medium (v1).
model, the refraction angle (r) increases as the Development of head waves is shown in Fig. 5.8.

Fig. 5.8 Development of surface


the head wave at critical
angle of incidence, ic for
(r = 90). Increase of the incident ic reflected
angle of refraction (r) with wave wave refracted
increase of angle of depth i waves
incidence (i)
M1, v1 interface
M2, v2 r headwave

90
96 5 Seismic Wave Transmission and Refraction

Head waves have great practical importance the direct wave. The sourcereceiver distance
as they form the fundamental basis of seismic (0 to xcr), at which the refraction wave overtakes
refraction exploration. the direct wave, in the same layering set-up, is
called the (cross-over distance).
The velocities of the two involved layers (v1 and
5.2.3 Ray-Path Geometry
v2) are computable from the slopes of the travel-time
and Travel-Time Curves
lines (as seen from the gure above). The velocity is
the inverse of the slope of the time-distance line.
On seismic shooting records, three main wave-
arrivals are usually observed. These are the direct,
refracted, and reflected waves. Their travel-time 5.2.4 Refraction Travel-Time
functions are linear for the direct and refracted Function
arrivals, but hyperbolic for the reflected wave. At
the critical distance (xc), where the incidence is at A travel-time function of a moving wave
the critical angle (ic = sin1(v1/v2)), a head wave is expresses the mathematical relation between the
generated which is refracted to surface at an angle time covered by the moving wave and the
equal to the critical angle. The distance (0 to xc), no source-receiver distance. This is also depending
refraction arrivals exist, hence it is a (shadow zone) on the velocity values of the two involved media.
for refracted waves (Fig. 5.9). Two cases are presented here; these are cases of
For a given interface separating two constant- horizontal and dipping plane interfaces.
velocity media, the refraction travel-time function
is a straight line tangent to the reflection hyper- (i) Case of Horizontal Refractor
bola at the point where both of these waves arrive Consider rst a simple one-refracting horizontal
at the same time. The distance at which reflection interface separating two layers of velocities (V1
time equals the refraction time, which occurs at and V2) as shown in Fig. 5.10.
the critical angle, is called the critical distance. Referring to this gure, the refraction ray-path
Due to its shorter travel-path, the direct wave, consists of three segments, AB, BC, and CD. The
at small receiver distance, arrives earlier than the refraction travel time T(x) is given by
refraction wave. However, after some time, the AB/V1 + BC/V2 + CD/V1. That is:
faster refraction wave catches up and surpasses
Tx 2z=V1 cos ic x2z tan ic =V2

reflection curve or, (using sin ic = V1/V2 from Snells law and the
Tx refraction curve sine-cosine relationship)
Tx x sin ic 2z cos ic =V1
direct curve
or,
 1=2
T0 Tx x=V2 2z V22 V21 =V1 V2
x
0 xc xcr where (ic) is the critical angle, and (z) is depth of
the interface. The two velocities, (V1 and V2) are
velocities of the two layers. These are equal to
z ic ic the reciprocals of the slopes of the direct and
M1, v1
refracted travel-time curves respectively.
head wave M2, v2 It is readily noted that refraction travel-time
function is linear in (x), and its curve is straight
Fig. 5.9 Ray-paths and travel-time curves of direct,
critically-refracted, and reflected waves, with critical line of slope equal to (1/V2) and its intercept (T0)
distance (0 to xc) and cross-over distance (0 to xcr) is given by:
5.2 Seismic Wave Refraction 97

Fig. 5.10 Ray-path x


geometry of a wave
A D
refracted from a horizontal
interface
Z
ic ic
V1
B V2 C

Fig. 5.11 Ray-path x


geometry of a wave A D
refracted from a dipping
z1 ic
interface

B ic z2

C V1
V2

 1=2
T0 2z V22 V21 =V1 V2 : z 1=2xcr V2 V1 =V2 V1 1=2

The depth (z) can be calculated from this


equation, where: (ii) Case of Dipping Refractor
As it is with the case of horizontal refractor, the
 1=2 ray-path in the case of dipping refractor (dip
z T0 V1 V2 =2 V22 V21 :
angle, h) consists of three segments; AB, BC,
Thus, with the velocity (V1) which is obtained and CD (Fig. 5.11).
from the travel-time of the direct wave, and (V2) The travel time T(x), in down-dip shooting, is
from the refraction travel-time, the interface given by AB/V1 + BC/V2 + CD/V1, where:
depth (z) can be calculated.
AB=V1 z1 =V1 cos ic ;
An alternative way of computing the depth
BC=V2 x cos h  z1 tan ic  z1 x sin h tan ic =V2 ;
(z) of the refractor is by use of the cross-over
CD=V1 z1 x sin h=V1 cos ic
distance (xcr). By denition, the travel-time of
the direct wave, at the cross-over distance, is
equal to that of the refracted. Thus, by equating (z1) is the perpendicular distance from the
the two travel times at the cross-over distance source at point (A) to the interface and (z2) is the
(xcr/V1 = xcr/V2 + 2z(V22 V21)1/2/V1V2), to perpendicular distance from the receiver at point
give: (D) to the interface.
98 5 Seismic Wave Transmission and Refraction

Hence, for down-dip shooting, the refraction apparent angle of dip in the direction of the
travel-time function is given by: shooting-line (x-axis direction).

Tx x sin ic h 2z1 cos ic =V1


5.2.5 The Delay-Time Concept
and for up-dip shooting, where (h) is given
negative sign, the function becomes: Travel-time functions that we have dealt with so
far are based on the assumptions that the inter-
Tx x sin ic  h 2z1 cos ic =V1
faces are regular plane surfaces. In nature, how-
ever, this is not always the case. By application
The angle of dip (h) can be computed from
of these functions on the more realistic cases,
two refraction-experiments: one down-dip giving
where refractors are irregular and dipping sur-
a straight line having slope (Sd) and the other is
faces, inaccurate results may be obtained. The
up-dip which gives a straight line of slope (Su).
concept of (delay-time), introduced by Gardner
From the travel-time functions for the two
(1939), serves as a tool for more accurate inter-
experiments, we have:
pretation of refraction survey-data where the
Sd sinic h=V1 waves are refracted from irregular and dipping
refractors.
Su sinic  h=V1
The delay-time (dTx) is dened to be the
difference between the actual refraction
Hence,
travel-time (Ta), corrected for the depth and
velocity variations of the weathering zone, and
ic h sin1 V1 Sd and
travel time required to traverse the horizontal
ic  h sin1 V1 Su ; giving: distance between the source and the receiver at
the highest velocity encountered on the ray-path
  (Sheriff 2002). Referring to Fig. 5.12, the
h sin1 V1 Sd  sin1 V1 Su =2 delay-time is represented by the difference
between the time the wave takes in covering the
In order to calculate (h) from this formula, the actually travelled path (SBCR) and the distance
velocity (V1) must be known. This is obtained (SR = x) with the refractor velocity (V2).
from the slope of the direct wave which is always The delay time (dTx) can be mathematically
available in such experiment. The travel-time expressed as:
curve of the direct wave is straight line passing
through the origin (T(x) = x/V1) and its slope is dTx Ta x=V2
equal to the inverse of the wave propagation
velocity. We may further note that (h) is the

Fig. 5.12 The delay time


concept, case of a wave
x
refracted from an irregular S R
dipping interface
zR
zS C
V1
V2
B
BS CR
BC BS CR
5.2 Seismic Wave Refraction 99

In case of one horizontal refractor, the sufciently accurate; provided that the
refraction travel-time function, is given by: refractor-dip is less than about 10 (dip < 10).
h i1=2
Ta x=V2 2z V2 2 V1 2 =V1 V2
5.2.6 Refraction in a Multi-layer
Medium
That is:

Ta x=V2 T0 In case of multi-horizontal reflector, the refrac-


tion travel-time function is derived by applying
Hence, the same principles used in the case of a single
refractor that is Snells law is applied to deter-
Ta x=V2 dTx mine ray-path direction every time the wave
crosses an interface. The velocities of the tra-
This result shows that the delay time concept versed layers must be increasing with depth in
is applicable to horizontal refractors. In fact the order that a head wave is created at the last
application of the concept on irregular reflectors refractor in the traversed layered medium.
is based on the assumption that the refractor is For the general case where the medium is
nearly horizontal under both of the source and made up of N horizontal layers of thicknesses
receiver points. This means that more accurate zn (n = 1, 2, 3, , N 1) and increasing
results are obtained when the refractor is of mild velocities Vn (n = 1, 2, 3, , N), the refraction
undulations and of relatively small dip. The travel-time of a wave critically refracted at the
assumption is that the distance measured along base of the last layer (layer N 1), is given by
the refractor surface is approximately equal to its Nettleton (1940, p. 253):
projection distance onto the horizontal plane at
X
N1 h i1=2
the surface. The approximation (BC  BS CR) is Tx x=VN 2 Dzn VN 2 Vn 2 =Vn VN
shown in Fig. 5.12. n1
The delay time (dTx) is usually presented in
two separate components: delay-time associated Reference to Fig. 5.13, showing a 3-layer
with the source point (dTS) and delay-time model, the travel-time function, T(x), of the
associated with the receiver point (dTR). Thus wave refracted from the base of the third layer, is
we have: given by:
dTx dTS dTR h i1=2
Tx x=V4 2Dz1 V4 2 V1 2 =V1 V4
It should be noted that, when the refractor is h i1=2
2Dz2 V4 2 V2 2 =V2 V4
horizontal, each of the delay-time components
h i1=2
becomes equal to half of the intercept time that is: 2Dz3 V4 2 V3 2 = V3 V4

dTS dTR T0 =2

Where, (T0 = 2z[(V2)2 (V1)2]1/2/V1V2) is 5.2.7 Refraction in a Medium


the intercept time. of Linear Velocity
The main application of the concept is in the Variation
determination of the refractor depth. The depth of
the refractor at the source point (zS) and receiver A non-linear travel-time function arises in the
point (zR) can be computed if velocity informa- case of a refracted wave travelling down a
tion is available. According to Sheriff and Gel- medium, the velocity of which is function of
dart (1995, p. 439), computation results are depth. For the case where the velocity is linear
100 5 Seismic Wave Transmission and Refraction

Fig. 5.13 Ray-path


geometry and
slope = 1/V4
corresponding refraction
travel-time curves for three
horizontal refractors slope = 1/V3
Tx
slope = 1/V2

slope = 1/V1

surface
Z1
V1

Z2 V2 > V1

Z3
ic V3 > V2 ic

V4 > V3

function of depth, V(z) = a + bz, (a and b are consist of two parallel but displaced linear seg-
constants), then the travel-time function is given ments of slope equal to the inverse of the velocity of
by Nettleton (1940, pp. 257261) and Dobrin the faulted layer (slope = 1/V2). These two seg-
(1960, pp. 7778): ments correspond to the refracted arrivals from the
reflectors original surface and from the down-
Tx 2=bsinh1 bx=2a thrown surface respectively.
In reference to Fig. 5.15, the fault throw (Z)
The ray-path, in this case, is of circular shape of is calculated from the difference in intercept
radius (r = zm + a/b), where (zm) is the maximum times (T), and the angle (ic) which is equal to
depth of penetration which is given by: cos1(P/Z). Since P = T  V1, and (sin
h i1=2 ic = V1/V2), we can write:
zm x=22 a=b2 a=b h i1=2
DZ DT V1 V2 = V2 2 V1 2
Geometrical shape of ray-paths and travel-time
curve are shown in Fig. 5.14. For more accurate determination of the throw
(Z) and fault location, another reversed refrac-
tion survey is needed to be conducted.
5.2.8 Refraction from a Faulted
Refractor
5.2.9 Applications of Seismic Wave
Refraction surveying can be used to detect a fault Refraction
and determine its throw. Consider a two-layer
system where the high-velocity layer (velocity, V2) Refraction eld work involves laying out a
is vertically faulted (fault throw, Z) as shown in spread made up of sources and detector, that suite
Fig. 5.15. The travel-time curve of refraction recording both of the direct and refracted
experiment across the strike of the fault plane will wave-arrivals. Because of the relatively long
5.2 Seismic Wave Refraction 101

Fig. 5.14 Sketch of


ray-path geometry and
travel-time curve for a
medium of linearly T(x)
increasing velocity with
depth

distance
0 x

c1 c2 c3 line of centres
r1 a/b
earth surface
zm x
0

Fig. 5.15 Ray-path slope = 1/V2


T
geometry and
corresponding refraction
travel-time curves for
refraction across a faulted
horizontal refractor
slope = 1/V2
T
slope = 1/V1

surface

Z1
ic Z2

fault throw, Z i P
c V1
V2

travel-path, refraction arrivals suffer from atten- (ii) Small-scale refraction surveys
uation in both of the energy and high frequency (iii) Seismic fan shooting
contents. To lessen these adverse effects, an (iv) Seismic tomography
adequate increase in energy source and adequate
(i) Large-Scale Refraction Proling
detector frequency response are used.
Large-scale refraction can be used in exploring
The main elds of application of refraction
geological rock layering of regional dimensions.
waves are:
Source-receiver spreads can be employed to
(i) Large-scale refraction proling
survey lines of several kilometers to several tens
102 5 Seismic Wave Transmission and Refraction

of kilometers. In this case, a proling technique small-scale refraction is in determination of the


is followed where the line (of say 2030 km physical characteristics (thickness and seismic
length) is covered in a series of few-kilometer velocity) of the surface weathering zone. This
long segments. The source for the nearest seg- activity is often carried out by seismic crews
ment may be few kilograms of dynamite-charge, working in seismic reflection exploration.
which can be increased to several tens of kilo-
(iii) Seismic Fan Shooting
grams for the large offset spreads. The technique
This is one of the early techniques of using
is applied to map regional-scale geological lay-
refracted waves in locating anomalous velocity
ering of an area.
bodies, such as buried salt domes. The detection
Larger refraction-spreads and greater
points (receivers) are laid out in fan-like arrays at
source-energy were used, in the study of the
distances from a common source point. This is
Earth-crust. For this purpose, spreads of more
the reason for labeling the method by this name.
than 500 km, using several tons of dynamite for
More than one fan-spread is usually applied in
the source energy have been used. Irrespective of
the one survey. A sketch of two-fan surveys is
the scale-size of the refraction-based exploration,
shown in Fig. 5.16.
the basic principles of the refraction method
Those ray-paths penetrating the anomalous
used, is the same. The nal result is linear event
body will show abnormal travel-times compared
from which the velocity and thickness informa-
with the others which are missing that body.
tion are obtained.
In case of a high-velocity body (as a salt
(ii) Small-Scale Refraction Surveys dome), a time lead of the arrival time of the
Normally, a small spread (100150 m) is received refracted wave will be observed. Thus,
employed. Depending on the shot position with by plotting the arrival times against the
respect to the detectors, there are two types of source-receiver distance, the abnormal values
spread; one is centre-spread type and the other is compared with a normal curve, delineation of
end-on like arrangement. Typically, detectors in the anomalous body is determined. For refer-
these spreads are laid down at 510 m spacing. encing purposes, the normal time-distance curve
This type of surveys is usually used in geo- is obtained, in a near-by location where there are
physical engineering purposes, as in investigating no anomalous bodies. This reference curve is
local geological changes. A typical application of obtained by plotting refraction arrival times

Fig. 5.16 Source-receiver Receiver


lay out used in seismic fan
point anomalous
shooting
body as:
salt dome

A
source
points
B
5.2 Seismic Wave Refraction 103

Fig. 5.17 Seismic source-borehole receiver-borehole


tomography, by
borehole-to-borehole
observation. All the
involved ray-paths and a
traversed anomalous body
are shown

source-point

receiver-point

against source-receiver distances. For more the surface. Like any other inversion process,
accurate mapping of the anomalous body more seismic tomography aims at determination, as
than one source point is used for the same accurately as possible, of the earth structural
receiver array. This is similar to the scheme used model for the area under study.
by the modern tomography procedure followed Seismic transmission tomography deals with
in exploring anomalies by source-receiver seismic transmission waves which, in general,
recording systems. More detailed description of experience refraction processes at boundaries
the method is presented in the geophysical liter- where velocity changes. There are two types of
ature like Nettleton (1940), Bth (1971) and seismic tomography, reflection and transmission.
Kearey et al. (2002, p. 116). Reflection tomography deals with reflection
travel-time measurements, while transmission
(iv) Seismic Tomography
tomography deals with source-to-receiver
The word (tomography) is of Greek-language
ray-paths along which the seismic waves are
origin, meaning (cross-section drawing). In the
directly transmitted. The more commonly
seismic eld, tomography found its way as an
applied technique is the transmission,
effective exploration tool under the name of
borehole-to-borehole (also called cross-hole)
Seismic Tomography. This technique has been
technique. In this method (cross-hole method), a
applied successfully in imaging seismic velocity
set of source points are distributed down a
variation and producing subsurface geological
borehole and another set of receivers are located
models. Seismic tomography is considered to be
in the other borehole, as shown in Fig. 5.17.
one of the modern techniques used in seismic
Another form of the seismic tomography is
exploration.
the surface-to-borehole technique. In this method
Seismic tomography is a type of inverse
the source points are located on the surface and
modeling (inversion process) which uses
the receiver points are distributed down the
acquired seismic data to generate earth models
borehole, as shown in Fig. 5.18.
(usually in terms of velocity and dimension
Interpretation of the data recorded from the
information). The more applied technique in this
cross-hole surveying (or from that of the
eld is the (transmission tomography) which
surface-borehole) aims at velocity distribution
requires putting the source points in a borehole
in the medium between the source and receiver
and the receiver points at another borehole or at
104 5 Seismic Wave Transmission and Refraction

Fig. 5.18 Denition of receiver-borehole surface


seismic tomography, using
the method of
surface-to-borehole
observations. All the
involved ray-paths and a
traversed anomalous body
are shown

source-point

receiver-point

sets. Iterative, forward modeling is usually cess, until computed travel-times become
applied with the help of an especially-designed closest to the measured travel-times. The end
tomography-algorithm, starting with an objective of the whole process is to determine
assumed model of velocity distribution. The the subsurface geological model as accurately
model is changed after each comparison pro- as possible.
2D Seismic Reflection Surveying
6

The seismic reflection method depends on the produces a data volume in which the seismic
same principle of the well-known echo phe- amplitude is function of the variables (x, y, & z).
nomenon. A mechanical shock generates a seis- Reflected waves carry useful exploration
mic wave (normally, P-wave) which propagates information in two forms: the travel-time and
through the earth material. On hitting a surface of waveform changes. Both of these two types of
a rock layer (an interface), the incident changes are functions of the physical properties
wave-energy is partly transmitted (passing and layering geometry of the subsurface geology.
through the interface), and partly reflected back In other words, the seismic information furnished
to the earth free surface. by reflection surveying, can supply two main types
The earliest successful application of seismic of exploration knowledge which are: structural
reflection technique was made about 1930 (Net- exploration information from travel-time mea-
tleton 1940, p. 233). Since then, and due to its surements, and stratigraphic exploration informa-
great efciency in the exploration work, the tion from wave-form measurements. These two
method was intensively applied in exploring types of changes constitute the principal seismic
subsurface oil-traps. In its conventional form, the tools for exploring the subsurface structural and
technique is based on initiating of a seismic wave stratigraphic geology (Fig. 6.1).
and recording its reflection arrivals by detectors
deployed on the earth surface along a straight
line. The normally applied technique, the seismic 6.1 Reflection Surveying Concepts
reflection surveying, is based upon the use of a
linear source-receiver spread moving along a 6.1.1 The Spread Configuration
dened seismic line. To cover a complete survey
of an area, the process is repeated for all lines of The shooting spread used in seismic surveying is
a dened line-network designed for that area. dened as the geometrical relationship between
The end result of this technique is obtaining a the location of the source-point and that of the
two-dimensional seismic section for each seismic receiver-array. This shot-receiver conguration is
line, which is, in effect, expressing the variation kept unchanged during surveying of the seismic
of seismic-reflection amplitude as function of the line. The main elements of a spread are the
two variables: distance, x and depth, z. The source-point (commonly known as the
section lies in the vertical plane that coincides shot-point), and number and inter-spacing of the
with the surface survey-line. For this reason, the receiver-points. In practice shot-points and
method is called two-dimensional 2D surveying receiver-points are not points but sets of points.
to differentiate it from the 3D surveying which Thus a receiver point is usually made up of a

Springer International Publishing Switzerland 2017 105


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_6
106 6 2D Seismic Reflection Surveying

Fig. 6.1 The two types of


knowledge (in addition to SEISMIC REFLECTION EXPLORATION
propagation velocity)
obtainable from seismic
reflection surveying;
travel-time for structural TRAVEL-TIME WAVE-FORM
exploration and waveform DATA DATA
for stratigraphic
exploration
WAVE PROPAGATION
VELOCITY

STRUCTURAL STRATIGRAPHIC
EXPLORATION EXPLORATION

group of detectors (geophones) and the shot-point multiple of the number (12). In 2D seismic sur-
consists of a group of source-points. That is why veying the number of spread-channels used has
these are commonly referred to as geophone increased from 48 in the early 1970s to 240; and
group and shot pattern. For computation pur- 360 channels in the following forty years. In 3D
poses, their geometrical centres are used to rep- surveying the number may exceed 1200.
resent the shot-point and receiver-points locations
respectively. The shape of the shooting spread and
6.1.2 The Seismic Trace
its main elements are shown in Fig. 6.2.
Based on the position of the shot-point in
The seismic trace, considered to be the building
relation to the receiver-point, there are three types
brick for the whole seismic survey data, is a graph
of spread conguration in common use. The
representing the recorded amplitude variation as
shot-point may be located at one end of the
function of reflection travel-time. The trace con-
receiver linear array giving a type called (end-on
tains all the outputs of an active seismic channel
spread). Another type of spread is the one in which
during the recording time. It is, in fact, a record of
the shot-point is located within the spread it is
the whole sequence of the reflection arrivals as well
called (split- or unbalanced-spread), and when it is
as other arrivals such as direct waves, refracted
located exactly at the centre of the receiver
head waves in addition to the various types of
array-line, it is called (centre spread). The three
seismic and none-seismic noises. There are three
common spread types are shown in Fig. 6.3.
common modes of trace-display. These are:
The number of the active receiver points (or
seismic channels) per spread is customarily xed (i) Wiggly-trace, when the amplitude variation
at a whole-number which is customarily taken as is represented by a wiggly line recorded.

Fig. 6.2 Geometrical Receiver Array


shape of the shooting
spread
Shot-point
offset spacing

Shot pattern Geophone group


6.1 Reflection Surveying Concepts 107

Shot-point
(i) end-on spread

Shot-point
(ii) unbalanced spread

Shot-point
(iii) centre spread

Fig. 6.3 Types of commonly used shooting spreads

(ii) Variable-area trace, when the peak-parts reflection point (RP) or common depth point
of the wiggly trace are blacked in. (CDP). Referring to Fig. 6.5, a seven-channel
(iii) Variable-density trace, when the ampli- spread (end-on type) produced a shot record
tude variation is represented by an appro- containing seven traces that have been reflected
priate shade-intensity. from corresponding seven CDPs.
An actual record of a 48-trace shot-point
Usually, the eld shot-records are displayed in
obtained from a centre-spread is given here-below
wiggly-trace mode (Fig. 6.4).
Fig. 6.6. It is showing fairly clearly, reflection
Seismic stack sections are displayed by super-
arrivals as well as direct and refracted arrivals.
imposing the wiggly- and variable-area modes of
display.
6.1.4 The Seismic Profiling
6.1.3 The Shotpoint Technique

A seismic survey of a certain seismic line starts Seismic surveying along a linear track is carried
with deployment of the spread elements (shot out by a technique known as (multi-channel
and receivers) along that line. The designed proling technique). The survey procedure
spread, which consists of linear array of involves the use of a xed-shape spread that
receiver-points which are co-linear with the moves along a linear course at a regular move-up
shot-point, is laid down along the line to be shift. The rst shot-record is obtained from the
surveyed. When activating the energy source (as spread which is laid out at the start-point of the
ring a dynamite charge), a seismic wave-front line. In the following step, the second shot is
advances through the medium in the form of recorded after shifting the spread location by a
spherical surfaces (spherical when the medium is certain step-up distance. This process
homogenous and isotropic). The seismic rays are (shift-and-shoot process) is repeated from start
reflected from intervening interfaces, back to the point to the end point of the line. The move-up
surface to be detected by the geophone groups distance is normally made to be integral multiple
planted on the surface. The reflection arrivals are of the receiver-point spacing. A map of all
then fed into the recording system via the elec- locations that have been occupied in surveying
trical channels which are displayed in a the entire line is called (surface-coverage map).
trace-gather (the shot-gather). For more clarication of the proling tech-
Each recorded trace (provided by an active nique, an example is given in Fig. 6.7 which
channel of the spread) represents a series of wave shows a seismic line of length of 39 receiver
arrivals of signals reflected from a subsurface stations, surveyed by an end-on spread. The
108 6 2D Seismic Reflection Surveying

Fig. 6.4 Part of 48-channel shot-record showing reflection arrivals and random seismic noise. Traces are displayed in
wiggly-trace mode

surface coverage, shown in Fig. 6.7, consists of midpoint. This means that each receiver-point
seven positions of the spread that advanced along of a spread will receive a seismic arrival reflected
the line at move-up shifts of four from a CDP located below the source-receiver
receiver-stations per shift. midpoint. In fact, shooting of one spread, the
Because this procedure gives a seismic section reflector will be sampled by a number of reflec-
representing a prole of the earth along the sur- tion points (CDPs) equal to the number of the
veyed seismic line, it is referred to as a proling spread active receivers. From the geometry of the
technique. spread ray-path, the CDP spacing will be half
that of the receiver points of the spread. Thus, the
subsurface coverage (CDP-line) covered from
6.1.5 The Fold of Coverage shooting one shot will be about half the spread
length (Fig. 6.8).
By denition the CDP is a subsurface point Seismic proling procedure is accomplished
located vertically below the source-receiver by repeating the shooting process (shot ring and
6.1 Reflection Surveying Concepts 109

R1 R2 R3 R4 R5 R6 R7
S

Reflector-1
CDP1 CDP3 CDP5 CDP7

TR1 TR3 TR5 TR7

TX

Reflector-1
Reflector-2

Reflector-3
reflection record

Fig. 6.5 Shot-point ray-paths and its corresponding seismic shot record. Spread (end-on type), made up of 7
receiver-points (R1, , R7) producing 7 seismic traces (TR1, , TR7) reflected from 7 CDPs (CDP1, , CDP7)

reflection-arrivals recording) at a series of equally it is shown that when the spread move-up is too
spaced shot-point locations along the surface line. large (larger than half the spread), the subsurface
By making the spread move-up distance equal to coverage is incomplete, with gaps in the CDP
half of the spread length, one reflection per one sequence where patches of the reflector are giv-
CDP will result for the entire line. The resulting ing no-reflection information Fig. 6.9a. When the
seismic section in this case is commonly known 12-channel spread is made to advance by
as (single-fold section). For larger move-ups, 6-receiver station move-up rate, the reflector is
gaps will result in the sampled reflector. That is, completely covered with single-fold coverage
certain CDPs will not have reflections at all. On (Fig. 6.9b). In the third case (Fig. 6.9c), the
the other hand, if the move-up distance is smaller spread is made to move with a spread move-up
than half the spread length, then CDPs will get distance of two receiver-stations, which resulted
more than one reflection, giving the case of in threefold coverage for the whole line except
multi-fold proling technique. The number of for the less-than threefold zones (called fold tails)
reflections realized per CDP is called the fold of at the beginning and at the end of the line. In this
coverage, (or just, fold). Of course the shorter the example, the fold, in the tail-zone, is building up
move-up, the greater the fold will be. from a single fold at the rst CDP to threefold
Considering an example of a seismic proling over 8 CDPs. The same thing occurs at the end of
case in which a twelve-channel spread is used. the line but in reverse direction.
The effect of the move-up distance on the fold of In a regular shooting, which is normally
coverage is claried in Fig. 6.9. In this example, applied in seismic proling, the fold of
110 6 2D Seismic Reflection Surveying

Fig. 6.6 An actual 48-trace shot record showing clear reflection arrivals with relatively high-amplitude direct and
refracted arrivals. Traces are displayed in variable-area mode of display

coverage (F) can be calculated from the fol- there is a direct relationship between fold and
lowing formula: reflection signal-quality appearing in the stack
trace. However the relation is not linear but fol-
F N=2n lows a power law of the form (F1/2) (Robinson
1983a, p. 111). Accordingly, the rate of change
where, (N) is the total number of the spread in the (S/N) enhancement diminishes with the
active receiver-channels and (n) is the move-up increase of fold (Fig. 6.10).
distance expressed in number of receiver-station
spacing covered in one move-up distance.
This equation shows that the fold is directly 6.1.6 The CDP Trace Gather
proportional to the number of channels and
inversely proportional with the move-up distance. As mentioned above, a shot-gather is the group of
The fold of coverage is a major factor in traces recorded by the spread channels of that shot.
increasing the signal-to-noise ratio (S/N). In fact Thus, a seven-channel shot has seven recorded
6.1 Reflection Surveying Concepts 111

receiver stations

1 10 20 30

source receivers

S1
S2
S3
S4
S5
S6
S7
Fig. 6.7 Surface coverage of a line made up of 7 spread move-up shifts

surface coverage (receiver line)

source
receivers

subsurface coverage (CDP line)

Fig. 6.8 Shooting spread surface-coverage, reflection ray-path, and CDP subsurface coverage, as realized from one
seismic shot-point

seismic traces. Likewise, the CDP-gather is a this regrouping process (trace re-sorting), a simple
group of seismic traces which belong to one proling example is given here (Fig. 6.11).
CDP. Naturally, the number of traces in a certain In this example (shown in Fig. 6.11) we have
CDP is equal to the fold of coverage in that CDP. a survey made up of ve shots (A, B, C, D, and
In a certain seismic proling process, the traces E). Shot-A of traces (a1, a2, , a6) and shot-B of
obtained for each shot (shot-gather), are re-grouped traces (b1, b2, , b6), and so on, for the rest of
into CDP gathers ready for being stacked after shots. The complete surface coverage is occu-
being subjected to certain processing steps which pying a total of 11 receiver-stations with one
aim at enhancing the reflection signal. To explain receiver-station for the spread move-up distance.
112 6 2D Seismic Reflection Surveying

(a)
move-up distance-1 move-up distance-2 move-up distance-3

S1 end-on spread S2 surface coverage S3 S4


gap gap gap
subsurface coverage
(b)
Move-up distance-1
Move-up distance-2
Move-up distance-3
Move-up distance-4

S1 S2 S3 S4

(c)
Move-up distance-1
Move-up distance-2
Move-up distance-3
Move-up distance-4
S1 surface coverage
S2
S3
S4
sub surface coverage

CDPs

FOLD 11112222 3333333333333333

Fig. 6.9 Surface and subsurface coverage in seismic Spread move distance is 2 receiver-stations, resulting in
proling surveying using an end-on spread. Move-up nominal fold of 3, with tail-fold at beginning and end of
distance: greater than half spread length (a), equal to half the line
spread length (b), and more than half spread length (c).

Corresponding to this surface coverage, is a known the (Coverage Diagram). In Fig. 6.11, the
subsurface coverage made up of 14 live CDPs, shot-gather is indicated by the horizontal arrow, and
(CDP1, CDP2, , CDP14). The trace gather for the CDP-gather is indicated by the vertical arrow.
CDP1 (CDPG-1) consists of one trace (a1). For Number of traces in each gather, represents the fold
the second CDP (CDPG-2) the gather has also of coverage. In this example, the fold is equal to 3
one trace (a2). The CDPG-3 has two traces for the range (CDP5CDP10), and tail fold for the
(a3 + b1), and so on. The number of traces per ranges (CDP1CDP4), and for (CDP11CDP14).
CDP is representing the fold of coverage. The re-distribution of the traces from the
Identifying the trace gathers (for shots and shot-gathers to CDP-gathers (called CDP sorting) is
CDPs) is much simplied by drawing what is shown in the following gure (Fig. 6.12).
6.2 Seismic Reflection Data Acquisition 113

seismic eld-crew usually carries out certain eld


activities directed towards determination of the
optimum survey parameters and optimum survey
procedures. Optimization of survey parameters and
S/N
survey procedures are collectively referred to as
(survey design) which is normally outlined in the
work plan normally prepared for the seismic survey.
fold of coverage, F

Fig. 6.10 Curve expressing type of dependence of the 6.2.1 Seismic Energy Sources
S/N ratio on fold of coverage of reflection signal. The
curve is of the power-law form (F1/2)
The traditional method applied in generating
seismic waves is exploding dynamite in shot-
holes. There are however, other methods which
6.2 Seismic Reflection Data have been introduced as alternative seismic energy
Acquisition sources. Choice of the source-type depends on the
surface conditions prevailing in the given survey
In seismic reflection data acquisition, we are, in
area. The main criteria considered in evaluating a
general, dealing with generating and detecting
particular source-type are the following:
P-waves which are propagating from the source
region and getting reflected from subsurface strong enough to generate strong seismic
interfaces. The complete data acquisition process signal that can be reflected from deep inter-
involves seismic source generation, reflected face and be detectable at the farthest receiver
wave detection, and digitally data recording. (receiver of maximum offset).
In order to record the reflection arrivals with the generated seismic signal is rich in high
minimum distortions, optimum operation parame- frequency components to be able to resolve
ters (called eld parameters) are determined and closely spaced reflectors.
applied in conducting the seismic survey. The the generated noise is of least energy level.

SHOTS receiver-stations
& 1 2 3 4 5 6 7 8 9 10 11
CDPS

SHOT- A SA a1 a2 a3 a4 a5 a6 Spread move-up direction


CDP- A a1 a2 a3 a4 a5 a6
SHOT-B SB b1 b2 b3 b4 b5 b6
CDP- B b1 b2 b3 b4 b5 b6
SHOT-C SC c1 c2 c3 c4 c5 c6
CDP-C c1 c2 c3 c4 c5 c6
SHOT-D SD d1 d2 d3 d4 d5 d6
CDP-D d1 d2 d3 d4 d5 d6
SHOT-E SE e1 e2 e3 e4 e5 e6
CDP-E e1 e2 e3 e4 e5 e6
CDP SEQ. 1 2 3 4 5 6 7 8 9 10 11 12 13 14

FOLD 1 1 2 2 3 3 3 3 3 3 2 2 1 1

Fig. 6.11 Example of proling process, where a horizontal arrow and the CDP-gather is indicated by the
6-channel end-on spread moving at one receiver-station vertical arrow. Nominal fold is 3 over the central 6 CDPs
move-up distance. Shot-gather is indicated by the and the tail-fold zones are at beginning and end of line
114 6 2D Seismic Reflection Surveying

called), in order to avoid high-frequency ltering


shot-point shot-trace gather
and seismic-pulse weakening caused by that
layer. To increase the charge-coupling with the
Shot-A a 1 + a2 + a3 + a4 + a 5 + a6
Shot-B b1 + b2 + b3 + b4+ b5 + b6 surrounding medium, the hole is lled with water
Shot-C c 1 + c2 + c3 + c4 + c 5 + c6 or with mud-water mixture (Fig. 6.13).
Shot-D d 1 + d2 + d3 + d4 + d 5 + d6 The most common way of using this method
Shot-E e 1 + e2 + e3 + e4 + e 5 + e6 is placing the dynamite charge at the bottom of a
drilled hole. However the charge, under certain
conditions, may be placed in shallow holes or
even at a height above the ground. A method
CDP RESORTING called Air-shooting or Poulter method (Poulter
1950), involves simultaneous ring of a number
of charges placed on poles in the air. This is not
CDP CDP-trace gather
an efcient method since the ground surface is of
CDP1 a1 high reflection coefcient and thus a relatively
CDP2 a2 small portion of the generated energy is trans-
CDP3 a3 + b1 mitted into the earth, and most of the energy is
CDP4 a4 + b2 reflected back from the surface and in generating
CDP5 a5 + b3 + c1 surface waves. This method (Poulter method) is
CDP6 a6 + b4 + c2 not in common use at present.
CDP7 b5 + c3 + d1 In general, the dynamite source is character-
CDP8 b6 + c4 + d2
ized by its high energy level, and of wide fre-
CDP9 c5 + d3 + e1
CDP10 c6 + d4 + e2 quency band, though it involves a certain degree
CDP11 d5 + e3 of danger and needs special storage conditions
CDP12 d6 + e4 and strict safety measures.
CDP13 e5 (ii) Weight Dropping
CDP14 e6 The Weight Dropping method (also called Geo-
graph or Thumper) involves dropping a weight of
Fig. 6.12 The CDP sorting process of the proling case, about 3 tons from a height of about 3 m on to the
given in the example shown in Fig. 6.11 ground (Fig. 6.14).
The seismic energy generated from the impact
of the falling mass with the ground is considered
Depending on the surface environments, the
to be weak in comparison with the dynamite
source types are classied into land- and
explosion. In order to strengthen the generated
marine-sources. Here-below are brief denitions
seismic signal, the dropping shot is repeated
of these seismic-energy types.
number of times (3060) shots done at the same
Land Seismic Sources location and electronically summed up to produce
(i) Dynamite one shot-record. This technique of summing
This is one of the most common types of seismic corresponding traces of shots of common location
source used in land seismic surveying. Normally, (called vertical stacking) is normally applied in
a certain dynamite charge is exploded inside a this and in other cases of weak energy sources.
drilled hole at a depth ranging from few meters to Compared with the other energy sources,
several tens of meters. Typically, the drilled hole weight dropping gives fairly high frequency, but
is of diameter of about 10 cm and depth of about less than dynamite-generated pulses. It is safe,
(515 m). Ideally the charge depth should be fast, and cheap in operation. The draw-back of
placed deeper than the base of the weathered the method is development of strong surface
layer (low-velocity layer, LVL, as it is often waves and, because of the ltering effect of the
6.2 Seismic Reflection Data Acquisition 115

Fig. 6.13 Dynamite


seismic-energy source
earth surface

LVL
weathering zone

dynamite charge

seismic rays

surface (low velocity layer, LVL), the high fre- the bottom of a heavy vehicle to increase coupling
quency components are severely attenuated. This of its plate with ground surface (Fig. 6.15).
method is rarely applied at present. On detonating the gas mixture, a sudden
pressure impact occurs which is transmitted
(iii) Gas Exploding
through the base-plate to the ground surface. By
This is another impulsive source (known as
hydraulic system the plate is locked into position
Dinoseis) in which the energy is created by
after the impact, to prevent repeated impacts.
exploding a mixture of oxygen and propane gases
Being a relatively weak-energy source, more
contained in a conned chamber, the bottom of
than three units are red simultaneously by a
which is a moveable plate resting on the ground
control signal sent from the recording system,
surface. The so designed chamber is attached to
and the shot is repeated many times, then vertical
stacking is applied to get improved S/N ratio.
The general features of Dinoseis are similar to
the Weight Drop method. Both are surface
sources which are generating relatively weak
falling mass seismic energy, developing strong surface waves,
and producing low-frequency seismic pulses.
The method is rarely applied these days.
Fig. 6.14 Weight-Dropping seismic-energy source

Fig. 6.15 Gas exploding


seismic-energy source

gas chamber
116 6 2D Seismic Reflection Surveying

(iv) Dynamite Cords Exploding in laying the explosive cords. Since the method
This seismic source, well known by the name depends on trenches for the explosive cords, it
(Geoflex), consists of an explosive cord which is means that it is not possible to be used in areas of
buried in the ground at a shallow depth (about too hard surface layer or in cities or villages. It is
half a meter depth). It is laid down by a not advisable to be used in areas where the sur-
hydraulically-operated plough which is espe- face layer is too thick, since in such conditions the
cially designed for this purpose (Fig. 6.16). transmitted energy will be much weakened.
Implementation of the Geoflex method starts
(v) Vibroseis
with laying the explosive cord (about 100 m
This is a non-impulsive seismic-energy source
length) inside a trench of about a half-meter
which was introduced in early 1950s and rapidly
depth and detonated at one end. While the
gained popularity as an acquisition tool. In 1982,
explosive-generated energy is travelling very fast
over 40 % of worldwide seismic surveys were
through the cord from the detonation point,
using it (McQuillin et al. 1984, p. 38). Unlike
seismic waves are generated from each point of
impulsive sources, Vibroseis method creates
the cord at delayed times. This will cause these
mechanical energy which is continuously vibrat-
waves to have phase differences bringing about
ing for certain time duration. A mechanical, truck-
the ltering effect similar to that made by
mounted vibrator is hydraulically driven, produce
source-arrays. Points of the earth surface coin-
electronically controlled vibration. The generated
ciding with the cord, are receiving the liberated
energy is conveyed to the ground by a metal pad
energy not at the same time, but in sequencial
(about 1 m2) attached beneath the truck
manner. For this reason, the down-going
(Fig. 6.17).
wave-front becomes inclined with respect to the
When in operation, the Vibroseis system
horizontal level. Along this wave-front (shown
transmits into the earth a seismic signal vibrating
by dotted lines in Fig. 6.16), highest energy level
at frequency which is varying linearly with time.
is obtained. In effect, there will be a ltering
This electronically-controlled vibration function
effect such that the near-surface horizontally
is called the Sweep. In practical application, the
travelling energy (surface waves) will be ltered
sweep duration (called sweep length) is normally
out before reaching detectors.
within the range (1020) s. Over this time-span
The main advantages of the Geoflex method, is
the sweep-frequency applied is (1050) Hz.
the attenuation of the surface waves, in addition
Apart from the taper imposed at both ends of the
to generating wide-band seismic waves. Regard-
sweep (about half a second in length), the
ing eld operation the method is fast and simple

Fig. 6.16 Geoflex


seismic-energy source. The
dotted lines represent the earth surface
in-phase generated waves
50 cm

explosive cord
Detonation
point wave-fronts
of
generated

Seismic waves
6.2 Seismic Reflection Data Acquisition 117

Fig. 6.17 Vibroseis


seismic energy source

vibrating
metal pad

amplitude is kept constant during operation time. The Sweep Removal


Further, the sweep may be started at low fre- The seismogram, y(t), obtained from exploration
quency then it is linearly increased (up-sweep) or using impulsive sources, is the result of convo-
started by high frequency (down-sweep). The lution of the source pulse s(t) by the reflectivity
up-sweep vibration is the more commonly function h(t) of the subsurface reflectors in
applied method. A schematic representation of addition to noise n(t), that is:
the sweep is shown in Fig. 6.18.
yt st  ht nt
As it is expected, the output reflection signals
of Vibroseis-generated energy are not short
The recorded seismogram is near-impulsive
wavelets, as normally seen with impulsive sour-
form if the source function is of an impulsive
ces. The reflection signal, in case of Vibrseis
form, as in case of dynamite source for example.
sources, is as long as the sweep-length used in
In Vibroseis shooting, the source function is not
the survey operation. In consequence, we get
an impulse, but a long time function represented
overlapping of the long reflection signals, and in
by the sweep. This means that the recorded seis-
this case, it will be very hard to distinguish
mogram is made up of overlapping sweeps which
individual reflection events on Vibroseis records.
is far from the familiar impulsive seismogram.
Comparison between outputs of vibroseis and
To convert the Vibroseis-generated seismo-
impulsive sources is shown in Fig. 6.19.
gram into impulsive ready-to-interpret form, a
The time between the end of the sweep, sent
signal compression process must be performed.
by the vibrator, and the end of the recording time
This process (called sweep removal) is done in
is referred to, as the (listening time).
processing stage by the cross-correlation process.

Fig. 6.18 Vibroseis (a)


sweep signals. a Up-sweep,
b Down-sweep,
amplitude
c Up-sweep with taper
applied at both ends time
(b)

amplitude
time
(c)
taper zone
amplitude
time
118 6 2D Seismic Reflection Surveying

Fig. 6.19 Ray-path (a) (b)


geometry and reflection source reflected source reflected
arrivals. a case of pulse pulses sweep sweeps
impulsive source and
b Vibroseis source where
reflected sweeps are
overlapping

In order to extract the impulsive-like record of is the (Unipulse), where the air continues
the reflections in their actual reflection-times, the flowing into the generated bubble for some
recorded traces are cross-correlated with the time after the initial discharge. This is done
applied sweep. This process can be done in the to lessen the effect of the sudden bubble
eld within the recording unit of the Vibroseis collapse which results in bubble oscillation.
system, or later by the processing software (see (ii) The Steam-Gun (also called Vaporchoc)
processing section). utilizes hot steam instead of air. It is
The main advantages of the Vibroseis method considered to be better than the Air-Gun,
are being fast, safe, and comparatively cheap to because with using hot steam, the effect of
run. It can be applied along roads and even in cities bubble oscillation is reduced.
since it causes no damaging effects on environ- (iii) The Gas-Gun is still another similar marine
ments. Technically, it has the advantage of having source which uses explosion of a mixture of
the source-function being under control. The propane and oxygen contained in a steel
sweep parameters (time duration, frequency range, chamber. The pressure pulse created by the
and taper) can be changed at will. The main explosion of the propane-oxygen mixture is
problem with the method is the low source passed to the water via a flexible seal which
energy-level. Like other weak surface sources, forms part of the chamber. The (Aquapulse)
vertical stacking is carried out at the same time as works on this principle.
the recording is going on. In normal surveying (iv) Dynamite explosion is implemented as a
work, several Vibroseis-trucks (typically 4) are marine energy source. A method called
operating at the same time, and about 2060 (Maxipulse) uses small dynamite charges
vibrations are conducted per each shot-point red directly in the water body. Another
location. The shot-gather traces per each shot, similar method (Aquaseis) uses long det-
are obtained from vertical stacking of all of the onating cord.
recorded traces involved in the one shot-point. (v) Electrical Arc Discharge is used to gen-
erate energy by electrical discharge in
Marine Seismic Sources water. Examples of this method are the
The most common types of seismic energy (Sparker) and the Wasp. This is a rela-
sources are: tively weak energy source which is more
(i) The Air-Gun is a widely used energy source suited for shallow-geology exploration.
in marine seismic surveying. It generates Out of these methods, the Air-Gun is by far the
energy by discharging highly compressed most commonly used type of energy source in
air into the water. A variation of this method
6.2 Seismic Reflection Data Acquisition 119

seismic marine surveying. In certain situations, the land and hydrophones for use in marine
Air Gun, is used as a land seismic source by gen- environments.
erating a seismic pulse from an Air-Gun sub-
(i) The Geophone
merged in a water-lled pit made on the earth
Detection of reflection arrivals in land surveying
surface. Typical example of such an application is
is done by geophones (seismometers as they are
in providing the seismic energy for VSP surveying.
sometimes called). The geophone is a
It should be noted here that the names of the
seismic-detection instrument which can trans-
energy types are trademarks given to them by
form the ground vibration motion into an elec-
their originators. The following Table 6.1 con-
trical voltage. The most commonly applied
tains a summary of the types of seismic sources,
geophones in seismic reflection surveys are the
with their trademark references.
electromagnetic type. It operates on the principle
of voltage generation in a coil moving within a
6.2.2 Seismic Detectors magnetic eld. The generated voltage is propor-
tional to the velocity of the motion of the coil
Earth movement due to the arrival of a seismic relative to the magnet. For this reason, geo-
wave is generally very small. The vibration phones of this type are normally referred to as
amplitude is of the order of 108 of an inch (velocity geophones).
(Gardner 1938). In addition to the The geophone is made up of a permanent
small-amplitude feature, the seismic waves have cylindrical magnet and a coil suspended by
two other criteria which the detection and leaf-springs inside a circular slit made in the mag-
recording equipments should be able to cope net. The slit is separating the magnetic south-pole
with. These are the wide amplitude-range and the (inner part of the magnet) from the north-pole
wide frequency-range. (outer part). The magnet is rmly attached to a case
The seismic detection process is based on which is tted with a spike for easy xing on the
conversion of the ground vibration to an electri- ground in an upright position (Fig. 6.20).
cal signal by a special transducer which can For operation, the geophone is planted verti-
respond to seismic amplitude and frequency full cally into the ground. In this position the magnet
ranges and without distortions. Seismic detectors will vibrate in vertical direction when the ground
are of two main types, the geophones for use in is seismically activated, and the coil stays

Table 6.1 Types of Land sources Trademark reference


seismic sources, with their
trademark references i Geograph Mandrrel Industries & McCollum Exploration Co.l
ii Dinoseis Sinclair Research Inc.
iii Geoflex Imperial Chemical Industries Ltd.
iv Vibroseis Continental Oil Co.
Marine sources
i Unipulse Petty-ray Geophysical Inc
ii Vaporchoc Compagnie Generale de Geophysique
iii Aquapulse Western Geophysical Co.
iv Aquaseis Imperial Chemical Industries Ltd.
v Maxipulse Western Geophysical Co.
vi Wasp Teledyne Exploration Co.
120 6 2D Seismic Reflection Surveying

Fig. 6.20 Schematic damping maximum damping applied without stopping the
representation of the resistor oscillation. Damping of about 0.7 times the
electromagnetic geo-
phone. Symbols N and critical damping gives a response which increa-
S denote magnetic north N S N ses smoothly with increase of frequency. This
pole and south pole damping factor (D = 0.7, in Fig. 6.21) is the
respectively damping which is normally applied in seismic
surveying environments.
(ii) The Hydrophone
This is a geophone-equivalent detector used in
magnet detecting seismic waves in marine surveying
environments. The hydrophone is a
pressure-sensitive detection device that uses
spike
substances that generate electrical voltage pro-
portional to pressure changes caused by the arri-
val of a seismic signal. Such a substance (called
piezoelectric substance) has a property of gener-
ating an electric voltage when subjected to pres-
sure. Piezoelectric transducers are also called
stationary because of its inertia. The generated electrostrictive devices (Sheriff 2002, p. 263). The
voltage in the coil is function of the vibration rate pressure changes, caused in a water medium, due
and the coil parameters (number of turns, radius, to passage of a seismic wave, are proportional to
and eld intensity of its magnet). the velocity of the water particles set into motion
In order to reduce oscillation, geophones are by the signal (Dobrin and Savit 1988, p. 63).
normally provided with a damping resistor In marine surveying, hydrophones are housed
shunted across the geophone terminals. The in a special hose lled with a certain liquid of
damping device controls the frequency-response such a density that makes the net density of the
of the geophone vibration. Typical forms of the hose (containing the hydrophones and liquid)
geophone response characteristic curves for approximately equal to that of water in order to
different damping factors, are shown in facilitate controlling its depth below water sur-
Fig. 6.21. face and in reducing effects of external sudden
Too heavy damping (over-damping) reduces forces affecting the streamer (cable jerking).
detection sensitivity, while too little damping Marine data acquisition is conducted by tow-
(under-damping) will lead to continued oscilla- ing the hose, equipped with hydrophones (called
tion. Critical damping is dened to be the streamer) behind the survey boat (Fig. 6.22).

Fig. 6.21 Schematic


representation of the
geophone frequency
response curves. Three
curves of three different D=0.3
Geophone D=0.5
damping factors (D) D=0.7
output

frequency
Natural
frequency
6.2 Seismic Reflection Data Acquisition 121

Fig. 6.22 Elements of the


transmitting
marine-surveying spread
tail buoy
(the streamer)

energy source

depressor depth hydrophones compass


controller

6.2.3 The Seismic Data Recording signal. With this unit the too-small amplitude is
increased and the too-large amplitude is reduced.
For about 30 years after seismic exploration was This operation results in compressing the signal
introduced, seismic shot-records were directly amplitude-range to that value which is matching
recorded on paper as wiggly traces. In the early the dynamic range of the recording unit. The
1950s, however, recording of seismic data on technique of amplitude matching with the
magnetic tapes was introduced in analogue recorder, for undistorted output, is an essential
mode. After about ten years, the more superior feature of all seismic recording systems. A mod-
digital recording technique was introduced. Soon ication of the AGC is the programmed gain
after mid 1965, the analogue method was com- control (PGC) with which the gain is not tied up
pletely superseded by the digital method. Despite to the input amplitude-variation. A PGC-coupled
the fact that analogue recording is now consid- amplier has its gain that varies with record time
ered to be obsolete, (that is, no more in appli- by a preset (programmed) time-dependant func-
cation), a brief description of the analogue tion. This is usually made available as an option,
system is presented here-below. This will be applied when it is required.
useful as the digital system involves some of the Next to the amplication part is the magnetic
components of the analogue system. recording system which produces a permanent
record of the data stored on magnetic tapes. This
(i) The Analogue Recording System
is accomplished either directly or after being
The analogue recording system carries out the
frequency-modulated (FM). A carrier frequency
recording in two stages: the amplication process
(typically, 4 kHz) is modulated by the signal
followed by the recording process. In the rst
received from the amplication system. In this
stage, the seismic signal, received from the
way it can produce a recorded signal of S/N ratio
geophone group, is amplied so that it will be at
up to 60 db, and frequency response which is
a level that suits the following recording stage.
practically flat over 3200 Hz (Evenden et al.
The seismic signal may be too small for the
1971).
recorder to detect, or too large for the recording
Through a playback of the recorded magnetic
unit to handle without being saturated. In order to
tape, or directly from the amplier output, a
cope with this wide range of amplitude variation
visual display may be obtained. The display unit
(estimated to be 80100 db) amplication factor
(camera unit) is usually included in the recording
is made to be variable. This is done by using a
system. This unit uses the principle of gal-
special electronic unit called the automatic gain
vanometer deflection caused by voltage variation.
control (AGC), which makes the amplier gain
By certain optical arrangement, galvanometer
to vary with the amplitude level of the incoming
122 6 2D Seismic Reflection Surveying

Fig. 6.23 Schematic magnetic tape


diagram showing the signal
flow in an analogue
recording system

pre-amp amplifier filters modulator tape


transp

magnetic tape
AGC camera recorder

geophone
group

deflections are recorded on photographic paper. The Analogue Part


The output is usually a wiggly trace record with The seismic signal outputted by the geophone
timing lines superimposed on it. A simplied group, which is in the form of an analogue electrical
block diagram of the complete analogue record- voltage, is transmitted through cable or by radio
ing system is shown in Fig. 6.23. waves, to the recording system. The incoming
A complete analogue seismic recorder is nor- signal is rst amplied by the (Pre-amplier) by
mally built, in such a way that each applying a certain gain-factor which is xed at a
geophone-group has its own complete recording constant value for all of the active channels. After
channel that consists of a complete this step, the signal passes to a group of analogue
amplier-to-modulator chain of units. This elec- lters. These lters, which are optional to apply,
tronic chain which starts from the geophone con- include (low-cut lter) for reducing low-frequency
nection and ending at the recording magnetic-head noise and (notch lter) for removing
is called a channel. Thus, a 24-channel station is mono-frequency interfering signals, such as the 50
equipped with 24 channels for seismic data and or 60 Hz signals picked up from high-tension
few more channels for recording the (time-break), power lines. Another lter which is always applied
the (timing signals), and (up-hole time). These is the (anti-alias lter). This a high-cut lter applied
additional channels are usually referred to as the to prevent frequency aliasing which occurs when
(auxiliary channels). frequency components are of higher frequency than
Nyquist frequency. Like all analogue lters, the
(ii) The Digital Recording System
anti-alias lter has non-zero phase characteristics.
The main feature of the digital recording instru-
This implies that each frequency component of the
ment is that it converts the incoming analogue
seismic signal will have its own time-delay. Con-
seismic signal to digital form. In this process the
sequently, certain wave-form distortions occur
signal is converted into a series of discrete values
whenever an analogue lter is applied.
at uniform time-intervals. This technique proved
to give higher delity recorded signal than ana- The Multiplexing Unit
logue method. In digital form, the seismic signal The multiplexer is an electronic device that makes
can be subjected to various mathematical analy- contacts between all of the active channels in turn,
ses using more sophisticated digital software. with the following digital part of the recording
The digital recording system can be divided into system. Effectively, the multiplexer reduces the
two main parts: the analogue-, and the digital parts, many analogue-data channels to a single channel
separated by the multiplexing unit (Fig. 6.24). that carry the data in a sampled form (Fig. 6.25).
6.2 Seismic Reflection Data Acquisition 123

analogue part multiplexer digital part

AMP FIL MAG.


TAPE
AMP FIL

AMP FIL TAPE


AMP A-D FR TR.

AMP FIL
AGC
AMP FIL GC
D-A

AMP FIL PAPER


REC. CAMERA

Fig. 6.24 Schematic diagram showing the signal flow in a digital recording system

d5 d4 d3 d2 d1
D
c5 c4 c3 c2 c1
C
b5 b4 b3 b2 b1 .. d3 c3 b3 a3 d2 c2 b2 a2 d1 c1 b1 a1
B Sample-1 Sample-2 Sample-1
a5 a4 a3 a2 a1
A

trace - sequential input sample - sequential output

multiplexing

Fig. 6.25 Principle of the multiplexer operation, trace-sequential input with sample-sequential output, using 4
channels (A, B, C, D)

To conceive the multiplexing operation, let us d1) in this sequence. In the second rotation cycle
consider a mechanical device in the form of an arm the outputted amplitude-values (samples) will be
rotating with constant rotation speed where the (a2, b2, c2, and d2). With continued rotation, the
arm head will trace a circular ring. Referring to output through a single channel will continue
Fig. 6.25, consider four channels, arranged feeding the A-D conversion unit, until the end of
equally spaced on that ring. In the rst rotation the recording time. In this way, the seismic data
cycle, the arm shall make contacts with the chan- has been converted from trace-sequential
nels (A, B, C, and D) in that order. In this cycle, the amplitude-values to sample-sequential mode, as
multiplexer will output amplitudes, (a1, b1, c1, and it is shown in this gure (Fig. 6.25).
124 6 2D Seismic Reflection Surveying

The Digital Part Next to the amplication system, comes the


The rst unit of the digital part of the system is analogue-to-digital unit (A-D converter), which
the gain-controlled amplier which keeps the performs two functions: measuring the
discrete sample-values, received from the multi- sample-value and expressing the measured val-
plexer, at an appropriate level, matching the dy- ues into binary words. Thus, each sample is
namic range of the A-D unit. In analogue expressed in binary word, normally of 14 data
recording system, gain control is applied con- bits (bit is 1 or 0) and a flag-bit denoting the
tinuously to the input signals which may undergo algebraic sign (1 for negative and 0 for positive).
gain changes as with application of automatic The measuring procedure is done electronically
gain control (AGC) or programmed gain control in a part of the A-D unit called (sample-and-hold
(PGC). The main disadvantage in that case, is circuit) by successive comparison of the sample
that the modications done to the seismic traces, value (voltage quantity) with pre-dened stan-
are permanent and the original signal level can- dard voltage-quantity.
not be recovered. The output of the (A-D) unit is a sequence of
This problem is solved by a special digital sample-values expressed in binary words which
amplication system called, (the binary flows into the next unit; the formatting unit (or
gain-ranging amplier) or (the digital gain formatter). Formatting process involves
amplier). This system applies the appropriate re-distribution of the digital bits (the ones and
gain and at the same time stores that applied zeros) of the binary words into a pre-dened
gain-values alongside the sample-time. In other arrangement with which the bit-position map will
words, the applied gain is recorded as a function be stored on the digital magnetic tape. In addition
of time which can be recovered at any time after to the seismic binary words re-formatting, other
recording is completed. With this provision, the related information is also made ready to pass with
applied gain and the original seismic traces can the seismic data to the tape transport unit. Seismic
be recovered from the recorded data whenever traces, auxiliary traces, tape- and trace-headers are
needed. The gain control unit (GC) is associated the data which this unit is organizing to be ready
with the digital amplication unit for the purpose for passing to the following unit. According to the
of keeping the sample value within the dynamic bit distribution made by the formatting unit, the
range of the following A-D conversion unit. sample values (expressed in binary words) are
The early recording systems, were provided recorded on the magnetic tape.
with this type of ampliers (called gain-ranging In addition to bit-location arrangement, the
ampliers) in which amplication values are formatting unit generates error-checking bits,
varied in 6-db steps rather than continuously as called (Parity Bit). For parity checking across the
in the case of the AGC units used in analogue recorded magnetic tape, the formatting unit pla-
recording system. In the early 1970, development ces a digit, of one or zero, in a dened track in
occurred in digital amplication, when a more the tape according to whether the total number of
advanced technique called (Instantaneous Float- ones (to be written across the tape at the sample
ing Point amplication) was introduced. These location) is even or odd respectively. In the
are fast (near instantaneous gain changes) and of play-back of the tape, if a bit is missing from the
higher dynamic range. The dynamic range is recorded tape, the parity bit will indicate this as a
dened to be the ratio of the strongest to the recording error. In case of a parity error, it is
weakest signals which can be recorded. The possible to exclude the erroneous binary word by
dynamic range of the analogue recording system a special command in a later stage.
is rated at about 40 db whereas the digital system The nal stage in the flow of the signal is the
(using binary words of 14 bits) has a dynamic tape drive (or the tape transport unit) which records
range of 84 db. the formatted data on magnetic tapes. About
6.2 Seismic Reflection Data Acquisition 125

twenty years ago, the tapes in common use were 21 When the cell is magnetized it represents (1) and
track-one inch tapes or 9 track-half inch tapes. In when it is not, then it is (0). This concept is claried
that recording system the rst track is usually from the following example (Fig. 6.26).
assigned for the parity check. At present narrower An analogue signal, shown at the top of
tapes (e.g. 4 mm tapes) and higher recording Fig. (6.26), is digitized and its sample values are
density tapes (in forms of small cartridges) were measured and converted into 8-bit binary words.
introduced for the seismic data recording. The eight bits of each sample (each binary word)
are recorded across the tape by magnetizing the
(iii) Concept of the Dynamic Range
cell when the bit is (one) and demagnetizing the
Any measuring or sensing instrument (geophone,
cell when it is (zero). The sign bit (marked as
amplier, recording system) has a bounded
S-bit) is a flag indicating the algebraic sign of the
capability of faithful detection. The system
sample value. It is assigned the value of (one) for
noise-level sets the detection lower limit, while
negative sign and (zero) for positive sign. For
the system technical characteristics set the
example, a sample value of minus nineteen
detection upper limit. For a given system, dy-
(19), the binary word in this example (9-track
namic range is dened to be the ratio (normally
tape) is (100010011), and for a value of (+27),
expressed in dbs) of maximum measured
say, will be (000011011), where the rst digit is
amplitude to the minimum recoverable signal,
for the sign (the sign bit).
where the minimum signal is taken to be the
In the case the recorded data is in multiplexed
noise amplitude-level.
form, the sequence followed will be in sample
The dynamic range (DR) is expressed by the
sequential, and if the data is in demultiplexed
ratio:
form, the recording will be in trace-sequential
DR A=adb mode. Normally, the magnetic tape is divided
into blocks where the rst block (called the
where, (A) is the maximum signal amplitude header block) carries general information like
which a system can measure (with tolerable record number, sampling period, record length,
distortion level) and (a) is the system noise-level. and other recording parameters. The next block is
The concept is often applied in evaluating the the seismic data which are usually in multiplexed
detection and measuring capabilities of recording form. Along with each sample value, the applied
systems. The early seismic analogue systems gain for that sample and sign-bit flag are recor-
which were recording directly on paper have ded. One track is reserved for the parity-check
dynamic range of about 25 db. The dynamic for checking the bit-column across the tape. The
range of analogue systems recording on magnetic end block contains information signaling the end
tapes is about 45 db. In case of digital recording of that set of recorded data.
systems, the dynamic range is dependent on the The Society of Exploration Geophysicists
number of bits used in expressing the sample (SEG) has developed several seismic data
value. For 14-bit recording, it is equal to 84 db. recording formats since the 1960s when the
digital recording system was introduced. In 1967,
(iv) The Magnetic Tape Recording the formats SEG-A, SEG-B, and SEG-X, were
The magnetic tape is a plastic strip, coated with published. Later on, other formats were pub-
magnetic material, is divided into tracks of cells. lished, as SEG-C in 1972, SEG-Y in 1975 and
After being digitized, each sample-value is repre- SEG-D in 1980. The formats of SEG-C and
sented by a binary word which is recorded on the SEG-Y are recording the data in de-multiplexed
magnetic tape by magnetization status of the cells. mode, whereas SEG-D can record in both
126 6 2D Seismic Reflection Surveying

analogue signal

(+)
0
(-)

sample values
6 9 22 69 66 15 0 -7 -4 30 32 11
0
2 0 1 0 1 0 1 0 1 0 0 0 1
21 1 0 1 0 1 1 0 1 0 1 0 1
22 1 0 1 1 0 1 0 1 1 1 0 0
3
2 0 1 0 0 0 1 0 0 0 1 0 1
24 0 0 1 0 0 0 0 0 0 1 0 0
5
2 0 0 0 0 0 0 0 0 0 0 1 0
6 0
2 0 0 0 1 1 0 0 0 0 0 0
2 7 0 0 0 0 0 0 0 0 0 0 0 0
S 0 0 0 0 0 0 0 1 1 0 0 0
digital magnetic tape

Fig. 6.26 Simplied sketch showing the principle of 9-track tape recording format. The binary word for each sample
value consists of 8 bits plus the sign bit (S)

multiplexed and de-multiplexed modes. At pre- (D-A) converter, demultiplexer, and the camera
sent, eld recording is mostly done using unit which displays the shot trace-gather (usually
de-multiplexed SEG-D format. in wiggly-mode traces) on the computer monitor
Developments in the tape-recording tech- or on paper. A typical shot record is presented in
niques concerned mainly the packing density and Fig. 6.27.
storage capacity. The density was developed This gure shows about one second of data of
from 800, 1600, then 6250 byte per inch. The a shot record made up of 48 seismic channels
tapes used nowadays are of 8 or 4 mm width, plus two auxiliary channels (at the left side of the
contained in cartridges or cassettes. record): the time break indicating the zero-time
of the shot, and up-hole-time. Traces are dis-
played in wiggly mode of display.
6.2.4 Data Playback and Display

For quality control (QC) purposes, the end result 6.3 Seismic Noise Characterization
of the recording station (the shot trace-gather) and Attenuation
needs to be displayed. The playback process is
the reverse of the recording process. It is done As we mentioned in a previous discussion, all
through a complete playback system which non-reflection arrivals are considered noise which
consists of a digital AGC, a digital-to-analogue cause distortions and masking effects to the
6.3 Seismic Noise Characterization and Attenuation 127

type of receivers, two factors are in operation in


noise attenuation. These are the number of
100 detectors in the one receiver (for attenuating
random noise and enhancing the reflection sig-
nal) and lay-out geometry of the detectors to
attenuate the coherent noises which are made up
mainly of surface waves.
The reflection signal approaches the detection
elements, deployed on the surface, simultaneously
(in phase) and hence shall interfere constructively,
when outputs of the detectors are summed up
together. The detected noise, being random, shall
interfere destructively and hence gets attenuated.
This process can be achieved in the process of
electronic stacking that is when the detection
elements (geophones) are connected electroni-
cally to form one output channel (Fig. 6.28).
The seismic recording system is usually
equipped with ltering provisions which can be
applied to lter out noises based on frequency
characteristics of the dominant noise. The hori-
zontally moving surface waves (and other co-
Fig. 6.27 Seismic record playback of 48-channel shot herent noises), on the other hand, will be
record displayed in wiggly mode attenuated by summing wave amplitude-values
of different algebraic signs. Extent of attenuation
is dependent on array geometry as shall be
desired reflection arrival (seismic signal). The explained in the next paragraph.
coherent noise are non- reflection seismic events
that may appear on a seismic record. Examples of
these noises are surface-waves (ground roll), 6.3.2 Coherent Noise Attenuation
air-waves, or body waves (direct-, refracted-, and
multiples). These noises and interferences are Attenuation of coherent noise is normally carried
generated from the source seismic-energy. The out by special design of the receiver array con-
other type (incoherent noises) is mostly of guration. The ltering action is governed by the
non-source origin. They are attributed to natural arrays number of elements, and element spacing.
and articial activities such as microseisms, wind This is expressed in terms of a mathematical
blowing on land or on trees or from trafc or formula for the receiver array-parameters (num-
man-made jerking actions. Incoherent seismic ber of elements and element spacing) and the
noise, which is of random nature, forms the apparent wavelength of the targeted coherent
familiar seismic background of seismic noise. Any receiver array has its own response
shot-records. formula, called the array response characteristics
which control ltering efciency of the horizon-
tally moving surface waves (the ground roll and
6.3.1 Random Noise Attenuation
such like waves).
The multi-detector receiver has the ability to (i) The Array Response Concept
remove, or lessen the effects of, both of the An array is a term used for a group of detectors
coherent and the incoherent noises. With this (geophones) connected to one output channel, or
128 6 2D Seismic Reflection Surveying

surface wave reflection


& random noise wavelets

+ + +

receiver
output
receiver array

reflection wavelets

reflector

Fig. 6.28 Role of a multi-detector receiver in attenuating random (and coherent) noises and enhancing reflection signal
of a vertically arriving reflected wave

a group of shot holes of a shot-point pattern. source) is used. The array source-elements
Compared with a single-element receiver, a (shot-holes, for example) are detonated simulta-
multi-element receiver (receiver made up of neously generating body waves which reach the
many geophones), has the capability of attenu- detector array after being reflected from subsur-
ating horizontally moving waves in addition to face interfaces. At the same time, the
enhancing the signal-to-random noise ratio by a source-generated surface waves travel from the
factor depending on the square root of the source to the receiver. As it is with the receiver
number of the array elements. Efciency of the array, the nearly vertical incident body waves are
array in attenuating surface waves is expressed in phase and hence they interfere constructively
by a characteristic function called the array to give enhanced reflected signal. For attenuation
response, dened to be the ratio of the amplitude of the surface waves, the length of the source
of an array-output to that of the same number of array should be equal to the wavelength of the
the array-elements gathered together at one generated surface waves.
location (Sheriff and Geldart 1995, p. 247). This The response of any given array (having given
is expressed in the form of a mathematical number of elements with given element spacing)
function called (directional response function, or can be presented in the form of a curve known as
just directivity function). (response curve or directivity curve). The
The theory, on which the concept is based, is abscissa is the ratio of the wavelength of surface
the same whether it is applied to the shot waves (coherent noise) to element spacing, and
hole-pattern or to the receiver geophone array. the ordinate is the array output expressed as the
Instead of using a single source point, a source ratio of the array-output to that of the same
array (shot-hole pattern, in case of dynamite number of elements gathered together at one
6.3 Seismic Noise Characterization and Attenuation 129

location. The array response is normally emphasized that (k) is representing the apparent
expressed in db units. wavelength measured in the array direction.
The basic principle is that waves travelling in Typical response curves (directivity curves) of
near vertical direction are enhanced while those geophone linear-arrays are shown in Fig. 6.29.
travelling horizontally are attenuated. This tech- The characteristics of the array response are
nique has been borrowed from engineering work summarized as follows:
done in radio-antenna design (Dobrin and Savit
The response curve repeats itself at
1988, p. 99). The application of the principles in
(x/k = 1), that is at (x = k), so it is com-
geophone- and source-arrays was published as
pletely dened in the interval (x/k = 01).
early as mid-1950s (Lombardi 1955) and (Parr
The curve is symmetrical about the point
and Mayne 1955).
(x = k/2)
(ii) The Array Response Function It has (n) lobes (maxima), where the principal
Directivity curves express the array response to main lobe centered at (k = ) and another
the different wavelengths of horizontally travel- equal lobe (the rst alias lobe) centered at
ling surface waves. For a linear array consisting (k = x)
of n elements with x inter-spacing, the response It has (n 1) nodes (zero-response) at
function, R(b), is given by: k = nx, nx/2, nx/3, , n x/(n 1). At
these points the attenuation is innite, that is
Rb sin nb=sin b there is no output.
It has (n 2) side lobes found in the range
where, b = px/k, and k is the apparent wave- k = nx to nx/(n 11). This zone, called
length of the surface waves travelling in the the (reject zone) or the (effective array
direction of the linear array. length), is the zone where the most effective
The R(b) function is periodic, repeating at a noise-attenuation occurs. Sometimes the
period of (b = p). The function is fully dened in reject zone is taken to be the zone found
the range of (b = 0 to p), that is, in the range between the 3 or 6 db points. A response
(x/k = 0 to 1). The curve, within this range, is curve for an 8 elements linear array is shown
symmetrical about the point (b = p/2), that is in Fig. 6.30.
about (x/k = 1/2), and it crosses the b-axis at
This gure shows that, the attenuation-zone
the points b = p/n, 2p/n, 3p/n, , (n 1)p/n.
(reject zone) lies between (k = 8x) and
The curve has a total of (n) maxima, where the
(k = 8x/7). This zone can be made wider (ex-
principal one is located at (b = 0), and the peaks
tended) and its response level lowered further, by
of the side lobes occur at the mid points between
increasing the number of elements (n). However
the zeros of the function.
this will cause narrowing of the pass band
Mathematical derivation is based on summing
(marked by the rst lobe where k  nx) and
of spatial harmonic waves moving along a linear
increasing its cutoff slope.
array made up of n elements, arranged at con-
The role of the directivity curves is in evalu-
stant interspacing x (Sheriff and Geldart 1995,
ation of array performances in attenuation of
p. 247).
horizontally moving surface waves. Given the
(iii) The Array Response Curve wavelength of the dominant coherent noise, a
The response curve is normally plotted as relative seismic detector array can be designed such that
response against (x/k), or against (k). The its response is most effective in attenuation of the
response value is plotted directly or in db units. noise having that wavelength. From the drawn
As it is mentioned above, the response, by de- curve, one can directly read the relative
nition, is expressed by the ratio between the array noise-cancelation effect corresponding to the
output and that of the same elements gathered at wavelength of the dominant surface waves (k), or
one location, that is when (x = 0). It should be to the parameter (x/k). This means that we need
130 6 2D Seismic Reflection Surveying

1.0 array response function


for

response
2-geophone linear array

array
0.5

geophone spacing/wavelength (x/)


0
1/2
0
1.0 array response function
for
response

4-geophone linear array


array

0.5

geophone spacing/wavelength (x/)


0
1/4 2/4 3/4

1.0 array response function


for
response

8-geophone linear array


array

0.5

geophone spacing/wavelength (x/)


0 1/8 2/8 3/8 4/8 5/8 6/8 7/8
0

Fig. 6.29 Schematic representation of response curves for different linear arrays

1.0

main alias
lobe reject zone lobe
0.5

R()

x/
0
0 1/8 2/8 3/8 4/8 5/8 6/8 7/8 1

8x 8x/2 8x/3 8x/4 8x/ 8x/6 8x/7 x

Fig. 6.30 Schematic representation of response curve for 8-element linear array, showing the main and alias lobes in
addition to the arrays reject zone
6.3 Seismic Noise Characterization and Attenuation 131

to know the wavelength of the surface-waves in The xed spread method is done by using
order to design the geophone array which is end-on spread which is made up of geophones
appropriate for the attenuation of the surface spaced at short distances (typically, 5-m spacing).
waves. The wavelength, which is representative After the rst shot is red and the shot record is
of this type of coherent noise, is normally produced, the shot is moved along the spread line
determined by special seismic experiment called by a distance equal to the spread length, and the
the (noise test). second shot-record is obtained. This
shift-and-record action is repeated a number of
times such that the resulting combined record will
6.4 Field Measures for Signal have its maximum offset to be about equal to the
Enhancement maximum offset planned for that survey. An
alternative method (called, the xed-shot method)
Before commencing a seismic reflection survey, is conducted by xing the shot location and move
certain eld procedures are normally carried out the receiver spread (along the spread line) away,
to determine the nature of the dominant noise in hence, the term (walk-away test).
the area in order to be able to design the source In both methods, each receiver point is
and receiver parameters which can output stron- occupied by a single geophone or by a group of
gest reflection signal and least possible noise geophones gathered in one location to give
level. For noise analysis, a special seismic clearest noise possible (Fig. 6.32).
experiment is done in the eld, called the noise The more commonly applied method is the
test (called also, walk-away or micro-spread xed-spread method because it is easier to move
test). Analysis of the resulting seismic record the shot location than moving the spread.
will lead to determination of coherent noise Noise-test operation is normally including
parameters such as apparent wavelength, veloc- measures for studying the broad-side noise as
ity, periods, and frequency, which are necessary well as the inline coherent noise. For this pur-
in the design process for both of the source and pose, another spread is laid perpendicular to the
receiver arrays. normal inline spread (Fig. 6.33).
The end result of the eld operations in con-
ducting the noise test is a noise-test record in which
6.4.1 The Noise Test the 5-m-spaced traces are drawn. It is important to
note that no ltering should be applied during the
There are two types of eld procedures that can recording process. For interpretation purposes, the
be followed to conduct the noise test in the eld. noise-record display should be produced without
These are the xed-spread method and the any type of time-variant scaling. Any amplitude
xed-shot method (Fig. 6.31). distorting process should be avoided in order to be

(a) fixed spread with moving shot


shot
spread
S1 S2 S3 S4 S5

(b) fixed shot with moving spread


spread

shot

Fig. 6.31 The two alternative methods used in noise-test surveying


132 6 2D Seismic Reflection Surveying

receiver points

Shotpoint

Fig. 6.32 Survey spread used in noise test surveying

cross-line
receiver
spread

inline receiver spread


Shot-point

geophones
grouped in one
location

Fig. 6.33 Noise test L-shaped spread to detect both of the inline and cross-line coherent noises. A receiver point may
be occupied by a single geophone or by a group of geophones

able to measure relative amplitudes, and for easier Interpretation of the noise-test data is nor-
event identication. However, time-invariant scal- mally done visually. It starts with identication
ing can be applied. Principal seismic noise events of the types of noises and other types of seismic
(and other wave arrivals) are schematically shown events. The main events a noise-test record
in Fig. 6.34. contains are: direct waves, refracted waves,

offset
A

B F
E
C
F

C
D time

noise cone

Fig. 6.34 Schematic representations of the main events, e back-scattered waves, f reflected waves. The group of
normally found in a noise-test record. a Direct wave, events between events (a) and (d) forms the noise cone
b refracted wave, c ground roll, d air wave,
6.4 Field Measures for Signal Enhancement 133

630m/s
2270m/s

reflection event

970m/s

450m/s

350m/s

Fig. 6.35 Actual noise-test record, showing reflection, refraction, sound, and surface waves

ground roll, air-wave (sound wave travelling period (s) are manually measured from the paper
through air), back-scattered waves, and wave record. From these parameters, frequency (f), and
arrivals of energy reflected from deep interfaces. wavelengths (k) are then calculated using the
The strongest events, appearing in a noise-test relations (f = 1/s, k = v/f). Interpretation results
record, consist of surface waves, covering a are found to be as quoted in the following
range of apparent velocities and apparent wave- Table 6.2.
lengths. These events make up the coherent noise These data show that the most distortive event
forming a group of large-amplitude events is the surface wave-3 (of velocity 970 m/s and
known as the (noise cone). An actual record of wavelength 108 m). It has the largest amplitude
noise test is shown in Fig. 6.35. and largest extent, covering all the record time
In interpreting the actual noise record, given and all of the offset range in this exampler. On
in Fig. 6.35, it is found that it contains body average the wavelength (100 m) can be consid-
waves (direct, refraction, and reflection events), ered as representative of the surface waves in the
surface waves (ground roll), and air wave (sound survey area, and can be adopted in the design of
wave). The wave parameters: velocity (v) and the geophone and shot-hole arrays.

Table 6.2 Values of Wave type V (m/s) f (Hz) k (m)


velocities, frequencies, and
wavelengths obtained from Sound wave 350 12 29
interpretation of the noise Surface wave-1 450 4 113
record shown in Fig. 6.35
Surface wave-2 630 8 79
Surface wave-3 970 9 108
Refraction wave 2270 25 91
134 6 2D Seismic Reflection Surveying

6.4.2 The Experimental Shooting done visually to determine the best parameters.
Choosing the appropriate parameter is based on
The source parameters depend on the type of the reflection-signal strength, frequency content,
source used to generate the seismic energy. For the reflection-signal resolution, noise level, and
dynamite source, there are three main parameters. extent of penetration depth.
These are: charge depth, charge weight, and
shot-hole pattern. The optimum charge depth and
charge weight are determined through direct 6.4.3 Determination of the LVL
experimental shooting. For the charge depth, Properties
several trial shots are conducted such that the
charge weight is xed while varying the depth. The earth surface is characterized by its
The same approach is followed in optimizing the non-uniform topography (variable elevation),
charge weight that is by xing the depth and and the surface layer is, in general, made up of
varying the weight of the charge Fig. 6.36. lose low-velocity material. This layer, (com-
As for the shot-hole pattern, the suitable monly referred to as the low-velocity layer, LVL)
number of shot-holes is determined by applying is made-up of one or more layers of velocity and
the same principles applied in the case of the thickness that can vary with location within the
receiver geophone array. However, very often survey area. The LVL thickness and velocity are
and for economical reasons, this is determined by typically of ranges (1050) m and (5001500)
the experimental-shooting method alongside with m/sec respectively.
the determination of optimum charge depth and Because of the large velocity contrast nor-
charge weight. Conducting multi-shot sources is mally found between the LVL material and that
another approach for increasing the of the medium below it, the base of the LVL acts
signal-to-noise ratio. This is normally done in as a strong reflector and refractor. Multiple
case of surface-sources as in the case of vibroseis reflections (reverberation and ghosts) can also
and weight dropping techniques. A typical plan develop in such environments. Effect of the LVL
for an experimental shooting for optimizing the is not restricted to the travel-time changes, but
source parameters is as shown in Table 6.3. also on the reflection waveform. In particular,
The shot records for these trial shots should be high-frequency components of the travelling
outputted with no ltering and with no seismic waves experience severe attenuation in
time-variant scaling being applied to them. the LVL due to the relatively strong absorption
A comparative study of the obtained records is phenomenon.

(a) charge depth test (b) charge weight test

Fig. 6.36 Experimental shooting for determination of: a charge depth and b charge weight
6.4 Field Measures for Signal Enhancement 135

Table 6.3 A typical plan for an experimental shooting for optimizing the source parameters
No. Shot-hole depth (m) Number of holes Charge weight (kg)/hole Shot-hole Spacing (m)
1 6 2 3 25.0
2 6 3 3 12.5
3 6 4 3 8.3
4 9 2 4 25.0
5 9 3 3 12.5
6 9 4 4 8.3
7 12 2 3 25.0
8 12 3 3 12.5
9 12 4 4 8.3

An integral part of the eld activity in a points arranged at certain depths inside the
seismic reflection survey is determination of the borehole Fig. 6.37.
properties of the surface layer, the LVL. In par- These dynamite capsules are red in sequence
ticular, effort is made to determine the thickness starting at the base of the hole and continuing
and velocity which are essential information upward till the last shot which is nearest to sur-
needed in correction-computations of the reflec- face. For recording the arrivals at the surface, a
tion and refraction travel times. The commonly group of geophones are usually planted at equal
applied methods are the up-hole surveying and distances from the surface location of the hole.
specially designed refraction surveying. For analysis and interpretation of the recorded
data, travel times (reduced to vertical travel-path)
(i) LVL Characterization by Up-hole
are plotted against depth. From the slope of the
Surveying
resulting plot, and the depths of the points at
The up-hole survey involves drilling a borehole
which slopes show abrupt changes, the velocity
of depth exceeding the expected LVL thickness,
and thickness of the LVL layer (or layers) are
normally within the range (50100 m). Small
calculated.
charges (dynamite capsules) are red at a series of

(a) (b)
shot hole detector vertical time, Tcos

depth LVL
(h) thickness

slant
shot time, T
depth
(h)

Fig. 6.37 Up-hole survey: a travel-path and b plot of vertical travel time (Tcosh) against depth (h), where (T) is the
slant time and (h) is the angle between travel-path and the borehole
136 6 2D Seismic Reflection Surveying

(ii) LVL Characterization by Refraction the reciprocal of the corresponding slopes.


Surveying Thickness (z) can be measured from its relation
Because the surface weathered-layer is usually of to the intercept time (ti), where, ti = 2z[(v2)2
velocity lower than that of the underlying bed- (v1)2]1/2/v1v2. The LVL parameters (thickness
rock, the refraction method lends itself as a tool and velocity) are computed from the data
for determination of its thickness and velocity. obtained from the two implemented shots.
For the same reason, the surface layer is normally
referred to as the low-velocity layer (LVL). The
refraction survey is conducted using a special 6.5 The Seismic Field Crew
short spread, typically (150250 m) provided
with 24 channels. Survey work of an area is carried out by a team
There are two alternative eld methods to of workers made up of professionally prepared
conduct the survey: xed-spread method, and personnel who are technically equipped to cope
xed-shotpoint method. In the rst method, two with all of the activities needed to complete the
shot-points are recorded, one on either end of the seismic survey. This is the seismic crew, or
xed-in-place spread. This is, in effect, two seismic party as it is sometimes called.
end-on shots are implemented for the xed A seismic crew consists of a number of sec-
spread. In the second method two shots, xed at tions each of which is specialized in one of the
the same point are red. For the rst shot, the eld survey activities. The main sections are:
spread is located on one side of the shot-point,
and for the second shot, the spread is moved to (i) Data Recording Section
the other side of the shot-point location. This is a This section is under the management of
spread set-up similar to centre-spread shooting. the Observer, who is responsible of the
The two methods are in Fig. 6.38. technical management of the recording
From travel-time curves, both of the velocity system (recording station) and magnetic
and thickness are computed from the slopes and tape recording. Under this section is the
time intercept of the produced curves. It is worth group of workers for geophone planting
noting here that the travel-times of both of the and control of the activities taking place
direct- and refracted-arrivals are linear functions over the seismic line during shooting. The
of distance. The velocity (v0) of the direct wave Observer usually submits complete daily
and that of the refracted wave (v1) are given by documented reports.

(a) (b)
time time time

distance distance

shot-1 shot-2 shots-1&2

direct arrival
refracted arrival

Fig. 6.38 Two methods of refraction survey specially designed for determination of the LVL properties (thickness and
velocity), a xed spread, and b xed shot-point method
6.5 The Seismic Field Crew 137

(ii) Topographic Surveying Section responsible for the dynamite storage and
Fixing on the ground, of the survey points transport taking all the safety and secu-
(shots and receivers) and measuring coor- rity precautions.
dinates (x, y, & z) of each of these points. (v) Mechanical Engineering Section
These data, in addition to data concerning For mechanical work needed by the crew
geographical nature and surface environ- as maintenance of drilling machines,
ments, are documented and reported. trucks, electricity generators and the
(iii) Drilling Section Vibroseis systems.
This is concerned with drilling the (vi) Administration and Finance Section
shot-holes, and other holes needed by the This section is responsible for personnel
survey, such as the deep holes needed for recruitment, living requirements, trans-
up-hole surveys. port, communications, material storage
(iv) Shooting Section and all nance affairs.
Workers in this section do all the nec-
These are the main sections of a typical crew
essary steps needed to prepare a
using dynamite for the seismic energy source. All
shot-hole. This involves preparation of
the sections are headed by the party chief who is
the right amount of charge and placing it
managing the crew work through direct contacts
at the required depth and lling the
with the crew and through the daily meetings
hole with water and mud mixture to
held every night with the sections heads.
secure coupling. The section is also
3D Seismic Reflection Surveying
7

In the years following the dramatic introduction three-dimensional shape, which is spherical, in a
of the analogue and digital data recording and homogeneous medium. In reality, the geological
processing, the main development-trends were in medium is made-up of different rock layers of
increasing recording channels and in more different physical properties and different geo-
advanced magnetic tape technology. These metrical shapes. With this realistic type of envi-
developments have led, in the early 1970s, to ronments, the advancing wave-front, is still
more effective exploration techniques including having the three dimensional form but no more
the introduction of seismic three dimensional perfectly spherical.
(3D) surveying. The rst proper 3D exploration The created seismic rays travel in every
work was conducted using xed cross spreads by direction in the space surrounding the source
Walton in the early 1970s (Walton 1972), and zone. In other words, the geological eld, as well
soon after the 3D seismic method entered appli- as the seismic eld in which it is created, is 3D in
cation on commercial-production basis. At pre- nature (Fig. 7.1).
sent the 3D seismic surveying became very In the conventional 2D shooting, the seismic
common especially for detail exploration and eld is created as three-dimensional wave-eld,
development of oil elds. but detected by two-dimensional array of
detection-points. Thus, the 2D surveying is
concerned with a limited portion of the reflected
7.1 Introduction 3D wave-energy. The rest of the reflected energy,
with all the useful information it is carrying, is
7.1.1 Nature Is Three Dimensional left to pass undetected.
In the 3D technique the source energy is uti-
The sub-surface geology targeted by seismic lized more efciently, since most of the reflected
exploration is, in essence, three-dimensional in 3D wave front will be detected. It is, in fact,
nature. The seismic eld created in seismic considered to be the more logical approach for
reflection surveys is likewise, three-dimensional seismic exploration than the conventional 2D,
(3D). In an ideal homogeneous medium, the because the 3D surveying conforms to the 3D
wave-front of the advancing seismic wave is of nature of the subsurface geology.

Springer International Publishing Switzerland 2017 139


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_7
140 7 3D Seismic Reflection Surveying

(a) Subsurface geology is 3D in nature of the transit-time with the depth of the well
which is also a 1D function.
In the conventional 2D seismic surveys,
where the shot-point and receiver-points are
co-linear, the resulting stack section consists of
a series of seismic traces each of which belongs to
a CMP location. The stack traces of a stack section
are uniformly spaced along the distance coordi-
nate (x). Thus the seismic amplitude in the pro-
(b) Seismic field is 3D in nature duced section is function of both trace-position
(x) on the seismic line and the two-way reflection
time (t). This means that the amplitude in the
produced seismic section, is represented by the
two-dimensional (2D) function f(x, t), or f(x, z)
when time is scaled by the propagation velocity.
With the 3D shooting technique, the
produced CMP-locations form a two-dimensional
array over the surveyed reflectors. In this case, we
Fig. 7.1 By nature, both of the subsurface geology have a seismic trace for each of these CMPs,
(a) and the seismic eld (b) created to explore it, are three
forming a data volume in which the seismic
dimensional
amplitude is represented as a function of its
7.1.2 1D-2D-3D Terminology position in space, dened by the three-
dimensions (x, y, and t) or by the dimensions (x,
A seismic trace can be viewed as a time function y, z) when the third dimension is expressed in
of the seismic amplitude which is normally rep- terms of depth, z. Thus, the amplitude is expressed
resenting variation of vibration-velocity as as three-dimensional (3D) function, f(x, y, t) or
function of recording time. If we disregard its f(x, y, z).
position information (that is neglecting the x-y In reference to (Fig. 7.2), the three terminol-
coordinates of its CMP location), the seismic ogy denitions are considering the seismic
trace is represented as a one-dimensional (1D) amplitude (a) as function of depth (z), function of
function of amplitude with time, f(t), or with (x, z), or function of (x, y, z).
depth, f(z). The synthetic seismogram is another The end product of the 3D seismic surveying
example of the 1D seismic-function. It represents is a seismic data volume which represents a
reflection-amplitude variation with the depth of three-dimensional function a(x, y, z) expressing
the drill-hole. The sonic log expresses variation the variation of seismic amplitude (a) with the

1D 2D 3D
y
x

a(z) a(x,z) a(x,y,z)

Fig. 7.2 Seismic data represented as 1D-function (seismic trace), as 2D-function (seismic section), or as 3D-function
(seismic data volume). The corresponding functions are a(z), a(x, z), a(x, y, z)
7.1 Introduction 141

three coordinates: x, y, and z. In this way, one (i) Erroneous Image Positioning
may consider the 2D section as a special case of Geometry of the 2D seismic line, assumes that
the more general 3D data-volume, in which one the source-receiver line, the incident seismic ray,
of the dimensions (x or y) is equal to zero. In the and the reflected ray are all lying in one vertical
geophysical literature it is customary to use the plane (the ray-path plane). Further, the reflection
terminology (data-box, or data-volume) for the point is located vertically below the
whole stack-data set outputted by the 3D survey. source-receiver mid-point. This is, in fact true
For display purposes, 2D sections from the data only when the reflector is a horizontal plane
volume, in both vertical and horizontal direc- surface or the seismic line is shot along the
tions, can be obtained. non-pinging axis of an anticline for example
(Fig. 7.3a). In the case of a dipping reflector,
however, the reflection events may be received
7.1.3 Limitations of 2D Seismic from points located outside the vertical plane
Surveying which is, in 2D surveying, assumed to be the
plane of the reflection ray-path (Fig. 7.3b).
Despite its outstanding success in subsurface A dry well may be obtained due to erroneous
exploration which has been achieved throughout positioning of the seismic image that may appear
the past years, the conventional 2D seismic on the 2D seismic section.
method is facing certain difculties in retaining An example showing erroneous image-
its well-established success standard. This is due positioning is the case of a seismic line passing
to the fact that most of the large structural traps near-by a dome. The 2D line does not cross the
with clear geophysical anomalies have already structural dome and yet the produced seismic
been discovered and what remained undiscov- section shows an anticlinal image (Fig. 7.4).
ered are those which are characterized by being Appearance of the image of a subsurface
small in size, complex in structure, weak in dome, which is not crossed by the vertical plane
seismic response, and may be situated in inac- containing the seismic line, is due to reflections
cessible areas. Limitations of the 2D seismic from the sides of the offside dome.c
surveying may be summarized as follows:
(ii) Distortion due to Dipping Reflectors
Strictly speaking, the 2D surveys give true struc-
A B tural pictures only when the surveyed areas are
made up by horizontally layered geology. In
RCVR line general, there is always a certain amount of
image distortion produced by 2D method when-
ever dipping reflectors exist. In processing of 2D
data, the reflection events are assumed to have
been received from reflection points which are
CMP line
located vertically below the surface seismic-line.
When the reflectors are not horizontal, and not
continuous, planes the 2D data will give erroneous
dipping subsurface images. With 2D migration, distortions
reflectors due to reflectors, dipping in the source-receiver
direction, can be corrected, whereas distortions
Fig. 7.3 Reflection ray-paths for zero-offset receivers in due to dipping in other directions cannot be cor-
2D surveying. a Line shot along strike direction, located
directly above an anticline of non-pinging axis. b Line
rected. Regarding dip-effect on seismic results we
shot parallel to line (a), but at horizontally-shifted position can distinguish the following three cases:
142 7 3D Seismic Reflection Surveying

reflection point offside dome

2D seismic line

time

2D seismic section

Fig. 7.4 2D seismic line shot and an offside dome. The produced seismic section shows a false structure due to
reflections from the side of the offside dome

R
S R S R

RP RP

RP
(a) (b) (c)

Fig. 7.5 Distortion effect of dip direction on reflection point location. a Case of no dip, b dip in the S-R direction, and
c dip is perpendicular to S-R direction. S and R are the source and receiver points respectively

(Case-1) Dip is of zero value (Fig. 7.5a) (Case-3) Dip has component perpendicular to
In this case, where the reflectors are the shot-receiver direction (Fig. 7.5c)
horizontal planes, the reflection point is The reflection point, in this case, is
located vertically below the shot- shifted outside the shot-receiver ver-
receiver midpoint, hence the reflec- tical plane. In this type of situation 3D
tion images, produced by 2D data, are survey has to be conducted and the
found in their proper positions, and no resulting data-volume must be sub-
distortions occur in this case. jected to 3D migration process in
(Case-2) Dip is in the shot-receiver direction order to get the correct subsurface
(Fig. 7.5b) structural image.
In this case, the reflection point is Distortion always occurs in case of dipping
shifted up-dip by an amount of shift reflectors and the distortion-severity depends on
which is dependent on the dip value. the dip and on dip direction. Thus, if a line in
The 2D method is adequate for this type 2D survey happened to be along the
of situation provided that the resulting dip-direction, then the reflection ray-path will be
stack section is corrected through in the source-receiver vertical plane, and in this
an appropriate 2D-migration process. case, no out-of-plane reflection points shall
7.1 Introduction 143

occur. The dip-caused distortion in this case, is On migration of the sections of these two
correctable in the processing stage (by applying intersecting lines, the strike-line section experi-
2D migration). In the general case, where the 2D ences no change whereas all events on the
line does not exactly coincide with the true dip dip-line section will be up-dip shifted. It is
direction, the reflection-events will not be cor- expected that this situation will lead to a mistie
rectly repositioned even with the process of 2D on the intersection point of the two sections. In
migration. In fact the amount of correction will fact, a stack section of a strike-line has two
be corresponding to that component of the dip problems. These are: creation of a mistie (at the
found along the line direction. To get the com- intersection-point) with the section of the dip
plete migration, the data must be subjected to line, and the misallocations of the reflection
another migration-process to be implemented in a CMPs which are not lying in the vertical plane as
direction perpendicular to the line direction the 2D stack sections are normally displaying.c
which is, of course, not possible. It can be con-
(iv) Weak Resolution of Structural Changes
cluded, therefore, that the 3D data volume,
Due to the wide spacing of the 2D lines, con-
obtained from a 3D survey, is a necessary
struction of the isochron contour maps are based
requirement in order to achieve proper migration
on interpolation of the reflection times measured
process regardless of the dip direction of the
on the survey lines. For this reason, structural
subsurface reflectors.c
details are generally not resolvable by the 2D
(iii) Creation of Misties surveying. Structural resolution gets worse as the
Let us consider the zero-offset sections (as stack line-spacing gets large compared with the sizes
sections) of two intersecting seismic lines of the structural anomalies. The isochrones map
(dip-line and strike-line) shot over a dipping of an area covered with ve 2D lines is shown in
plane reflector (Fig. 7.6). (Fig. 7.7a). In this case, the 2D survey failed to
For the strike line, the reflection ray-path resolve the three domes detected by the 3D sur-
plane is slant and not vertical, as it appears in the vey carried out in the same area (Fig. 7.7b).
seismic stack section, and the reflection events The weak resolution power of 2D surveying
obtained from the dipping reflector shows no becomes more serious problem when the struc-
dipping feature. On the other hand, the dip line tural anomalies are of dimensions smaller than
will be of slant ray-paths, all lying in the vertical the applied line-spacing.
ray-path plane.
7.1.4 Merits of the 3D Technique
dip line
surface plane
Due to economic constraints the 3D method is
strike line used only in cases when more detailed and more
zero-offset
reflection accurate exploration are aimed at. Typically, it is
ray-path
applied for restricted purposes such as develop-
ment and appraisal of already discovered
zero-offset oil-elds. When economy allows, however, 3D
reflection CMPs
ray-path P surveying is applied in the normal general-
Reflection
purpose seismic exploration.
dipping plane As we have seen above, the 2D method gives
CMPs
distorted images of the subsurface geology espe-
cially in case of dipping layers. The 2D method also
suffers from weakness in its ability of resolving
Fig. 7.6 Effect of dipping reflection plane on zero-offset
small geophysical anomalies and ne structural
reflection points in case of two 2D intersecting lines.
Strike stack section shows no dip, whereas dip line does. and stratigraphic details. Inability of surveying
On migration, a mistie occurs at intersection point (P) inaccessible areas is yet another limitation of
144 7 3D Seismic Reflection Surveying

Fig. 7.7 Sketch maps (a) (b)


showing isochrones
contour maps from: a 2D Line-1 Line-2 Line-3
survey and b from 3D
survey which is revealing
more structural details than Line-4
the 2D map

Line-5

the 2D seismic technique. These limitations were direction, in addition to possibility of getting
removed or very much reduced by the application horizontal sections (time-slices) at any level
of the 3D method. The advantages of the 3D sur- within the data-volume (Fig. 7.8).
veying may be grouped under the following three
(ii) Capability of Surveying Inaccessible
groups of application domains:
Areas
(i) In the Geophysical Domain As far as seismic surveying is concerned, an area
3D surveying is furnishing good control on the is considered to be inaccessible, when neither
determination of processing parameters, such as seismic energy-sources nor receivers are allowed
static values, stacking velocity, and dip-vectors, to be located within the area boundary. An
and hence it allows more effective three- inaccessible area can be surveyed by deploying
dimensional migration process. Increased ef- the receivers on the boundary of the area and
ciency in the use of seismic source energy, due to shooting at source-points which are also dis-
the large number of active detection channels tributed over the area boundary. The surveying is
normally used in 3D eld acquisition. The 3D data conducted by shooting the source points in
processing provides the stack data-volume, which sequence while the receiver spread is kept xed
allows displaying vertical sections at any throughout the shooting process. For each shot,

Fig. 7.8 Types of sections


which can be extracted
from the 3D data box

3D data box

vertical
y-sections
vertical vertical
x-sections sections

horizontal
sections
7.1 Introduction 145

Fig. 7.9 Concept of loop


shooting applied in 3D boundary line
surveying of inaccessible of the
areas. A CMP is created at
inaccessible area
the middle of each
source-receiver distance.
Receivers are deployed
around the inaccessible CMP receiver
area points

source
point

there will be a number of CMPs located at the (iii) In the Geological and Reservoir
sourceto-receiver midpoints, which will be Domains
located within the surveyed area (Fig. 7.9). The 3D data can furnish accurate information on
By this type of shooting (called loop- the subsurface geological structure of the area.
cshooting), the area will be covered with CMPs Unlike 2D data, it can resolve small and complex
without having survey points (sources or recei- structural and stratigraphic anomalies. It provides
vers) existing within the area. Special software direct information (with no interpolation proce-
can handle the recorded data and get it processed dure) of the subsurface geology expressed in
to produce a 3D data-volume. its real three-dimensional image. Due to the
Although this technique makes seismic sur- dense sampling points of the geological space,
veying of an isolated inaccessible area possible, the 3D data becomes more readily-interpretable,
it suffers from a number of weak points. The leading to increased degree of accuracy and
distribution of both of the recorded CMPs and much improved resolution-power (Fig. 7.10).
the fold of coverage are not uniform. Also, it is In the reservoir domain, 3D data provide
possible to lose shallow reflections with traces delineation of the 3D shape and spatial limits of
recorded at large-offset receivers. These limita- oil-elds as well as giving more accurate esti-
tions become less effective, the smaller the area mation of their hydrocarbon reserves. The data
and the deeper the targeted reflectors. Applica- can help in outlining oil/water contact which can
tion of the concept (loop spread surveing) was assist in more accurate emplacement of devel-
dealt with in more detail in (Alsadi 1992, 1994). opment and production wells. Reservoir studies,

Fig. 7.10 Sketch showing 3D seismic data box


that the 3D data provides 2D seismic sections
more dense sampling
points of the subsurface
geological space than the
2D data
146 7 3D Seismic Reflection Surveying

based on 3D data, help in more accurate reservoir study and prepare its nal report for guiding work
characterization in terms of facies distribution in the plan implementation.
and fluid content. Such information would lead The preplanning phase is an important pre-
to more accurate oil-eld characterization and requisite needed in order to facilitate the work
enhanced production rates. execution especially in the data-acquisition
stage. Pre-knowledge of the area assist the geo-
(iv) In the Economical Domain
physicist to be prepared, and make the necessary
Despite the overall high cost and long survey
provisions for any problem that may crop up
duration compared with the 2D method, the 3D
during the work. Examples of such problems are:
method is considered to be cost-effective when the
skip shots, make-up shots, seismic-energy to be
merits of its results are taken into consideration.
used, work permit, and other administration and
From the economical point of view, 3D data would
logistical matters.
give increasing success-ratio (producer-to-total
The preplanning report of a 3D seismic survey
wells ratio) and improving well production-rates.
normally covers the following basic items:
The over-all eld evaluation-cost is reduced
through less development-wells, proper well (i) Compilation and assessment of available
emplacement and shorter development-period. technical data.
In view of its exploration efciency, the 3D (ii) Survey objectives.
method gives an enhanced success ratio of (iii) Operational constraints.
development drilling and increase in the oil (iv) Survey-design plan.
reserves. These features form the basis for the (v) Scouting and preliminary tests.
justication of conducting 3D surveys especially (vi) Work organization and scheduling.
for oil-eld appraisal and eld development. In (vii) Cost analysis.
comparison with development drilling, the justi- (viii) Health, safety, and environment
cation is even stronger in comparison with the protection.
2D method. For example, a 10 km2 3D survey
It should be noted here, that the item of the
may cost no more than 10 % of that of drilling a
contract preparation is deliberately omitted from
well of 3000 m-depth. The overall saving
the above mentioned list. The omission of this item
achieved from a 3D survey is represented by
is made on the basis that it represents an indepen-
reduced eld-evaluation costs and by earlier
dent piece of work to be carried out in the following
production due to shortened development-period.
stage, rather than being part of the preplanning
report. This is natural procedure, since the contract
items are depending on the survey parameters
7.1.5 Survey Preplanning which need to be prepared and made ready before
starting of contract drafting. Here-below we shall
Although, in principle, work preplanning is give brief explanatory notes for each of the items
essential in any seismic surveying project, it included in the preplanning report.
becomes more necessary prerequisite when deal-
ing with 3D surveying. The urge for the preplan-
ning phase stems from the complex nature of the 7.1.6 Cost Considerations of 3D
3D surveying work which involves large nancial Surveys
investments. A feasibility study should be con-
ducted as a joint effort by a team of personnel from 3D surveying is usually quoted as being a costly
all parties concerned with the 3D survey project. In process. To be more objective, however, the cost
particular, the team should include specialists in the assessment must be made in relation to the
data acquisition, data processing and interpretation, turnover benets. In fact, even 2D surveying can
as well as experts on legal and contractual matters. become very expensive when using very small
The task of this team is to carry out the preplanning line spacing, and high fold of coverage. It is
7.1 Introduction 147

generally accepted now that the 2D and 3D the region of (1015) % of the cost of the data
methods become equivalent cost-wise when the acquisition. These gures are near-realistic when
2D lines are spaced by about half a kilometer. the survey-points for both methods are spaced by
The 3D surveying is considered to be 50 m and the coverage folds are 1600 and 3200 %
cost-effective compared with the 2D method for the 3D and 2D respectively. In general, the
considering the merits of the 3D data. Compar- cost of 3D surveying depends on the surface
ison between 2D and 3D survey-cost is more environment of the survey area (Fig. 7.11).
meaningful, when it is realized that 3D data
(ii) Dependence of Cost on Survey
provide dense spatial sampling of reflectors,
Parameters
enhanced structural resolution, and undistorted
The survey parameters which are most influential
3D images.
on survey cost are the spread parameters, fold of
The 3D-Survey Cost Elements coverage, bin size, and shot density used in a given
In discussing survey costs, and 3D-to-2D cost survey. These elements are interrelated in their cost
comparison, it is of course very difcult to quote contribution. Thus, fold (F) may be statistically
precise gures. However, it is possible to give a estimated from the formula:c
fair assessment of the main elements of the cost.
These elements are important issues to be con- F Ntr =Nbin
sidered in cost analysis of any 3D surveying
project. where, Ntr and Nbin are the survey total number
The main cost-elements of the 3D survey can of traces and total number of bins respectively.
be summarized as follows: By substituting the product (shot density,
SHD) x (survey Area) x (live channels per shot,
(i) Surface Conditions NCHN), for Ntr, and (survey area/Bin Size) for
Difculty in quoting precise gures is due to the Nbin, we get the important relation:
wide variation between types of surveys. In par-
ticular, prevailing surface conditions, and types of F SHD  NCHN  Bin Size
the applied survey parameters differ from survey
to survey. In the present world-wide seismic Hence,
industry, the cost of the data-acquisition of
onshore surveys (on-land surveys) is approxi- SHD F=NCHN  Bin Size
mately in the range of (510) k$/km for 2D
method and (2040) k$/km2 for 3D. The cost of Since total survey-work depends on shot
processing and interpretation is estimated to be in density (SHD), this formula is showing that the
total work done in conducting a 3D survey (and
cost hence total cost) is directly proportional to fold of
k$/km2 coverage and inversely proportional to bin size
and number of active channels used per shot.
100 A closely related to cost, is the (turnaround
time), or the duration of the complete 3D survey.
For a given survey (given area size and given
50 survey parameters), survey duration depends on
the work-production rate, which is, in turn,
farm, shallow Jungle dependant on the applied work parameters, like
offshore desert urban water survey parameters, type of energy source, surface
conditions, eld processing, topographic survey,
Fig. 7.11 Sketch showing dependence of survey cost permit fees, and number of work-shifts fol-
(cost of eld data acquisition) on surface environment lowed in conducting the survey.
148 7 3D Seismic Reflection Surveying

It is found that the eld data-acquisition of a subsurface circle will have a radius of half that of
land 3D survey of a 500 km2-area, say, would the surface circle, and all reflected waves received
last for about (23) months. Adding to this, time by the detectors, distributed over the surface cir-
needed for processing and interpretation, the total cle, will have arrived at the same time. However, if
duration time to complete a 3D project of such an the receiver points are laid down on a straight line
area is estimated to be in the order of about (68) passing through the source point, the reflection
months. points from which the incident energy is reflected
and detected by the co-linear receivers, will fall on
a subsurface straight-line with point-spacing equal
7.2 Definitions and Basic Principles to half that of the receive points. In this case, the
detected reflection information (seismic ampli-
7.2.1 Definition of 3D Surveying tude) is restricted to the 2D vertical plane con-
taining the reflaction ray-paths (Fig. 7.12).
Consider a case where seismic energy, generated As this gure is showing, the difference
at a surface point, is reflected from a subsurface between 2D survey and 3D survey is based on
horizontal planer reflector. The generated seismic the way the spread elements are distributed on
wave-front, which is of 3D form, spreads out the surface. In fact, what makes a survey 2D or
through the geological medium (assumed to be 3D is the way the receivers are deployed on the
homogenous) with a velocity decided by the surface. In 2D surveying, receivers are laid down
physical properties of the medium. At the instant in line with the source point, and the used spread
the wave-front hits the reflection plane, a circular is linear, made up of one-dimensional array of
zone of it (of the reflection plane) will be illu- receivers, and in this case, the reflection points
minated by the incident wave-front which is, will fall on a subsurface straight line which is the
according to reflection laws, reflected back to the projection of the receiver line onto the horizontal
surface. reflection-plane. In the case of 3D surveying, the
Using the ray-concept, detection points dis- spread consists of receivers which are distributed
tributed over a surface circle centered about the over an area, and the used spread consists of a
source point, will receive reflected energy from two-dimensional array of receivers. The reflec-
reflection points located on a corresponding sub- tion points in this case will be distributed over a
surface circle. Based on the laws of reflection, the subsurface area of the reflector-plane.

Fig. 7.12 3D survey is


R
when source and receiver
points are covering an area, 3D S R R R R
whereas 2D survey 2D
receivers are co-linear with
the source point

RP RP RP RP RP
subsurface reflector plane
7.2 Definitions and Basic Principles 149

7.2.2 The 3D Spread-Geometry 7.2.3 Concept of the Trace Azimuth

The spread in the 2D geometry consists of a Considering horizontal plane reflectors, all the
receiver points co-linear with the source point, reflection ray-paths, of the CDP trace-gather in
whereas the spread in 3D survey takes the geo- 2D surveying, are coincident on one common
metrical form of a rectangular grid of receiver vertical plane (the ray-path vertical plane). In the
points. Typical 2D and 3D spreads are shown in 3D method, however, each reflection ray-path
the following Fig. 7.13. may fall in its own vertical plane. These ray-path
Whether it is 2D or 3D survey, the planes are generally not coinciding on each other,
reflection-points (RPs), created from ring a because of the varying source-receiver directions
shot-point, are spaced by half the receiver spac- (azimuths) of the CDP gather-traces. Thus, the
ing. The spread used for 3D surveying (3D receiver azimuth in 2D surveying is constant. It is
spread) is normally covering a rectangular area. in the direction of the seismic line. The corre-
From geometrical consideration, it can be seen sponding azimuth, in 3D surveying, is variable
that the subsurface area covered by the reflection because of varying receiver bearing in relation to
points is also a rectangle which is of an area the source point.
equal to quarter of the area covered by the 3D The ray-path planes of the reflection seismic
spread. This can be readily observed from the rays (of a 3D-survey shot-point) are vertical
Fig. 7.13. planes but they assume different directions. This
Shot-records, obtained from 2D and 3D is the case because the receiver points are located
spreads, are shown schematically in the follow- at different bearings in relation to the shot loca-
ing Fig. 7.14 tion. The ray-path bearing, with respect to the

Fig. 7.13 Type of spread (a)


decides whether the survey R R S R R
is 2D or 3D.
a One-dimensional array
spread for 2D survey and
b two-dimensional array
spread for 3D survey

RP RP RP RP RP

R R R R R
(b) S

RP RP RP RP RP
S RP
source receiver reflection point
150 7 3D Seismic Reflection Surveying

Fig. 7.14 Sketch showing (a) (b) line-1 areal spread


2D and 3D shot-records. linear spread
line-2
a 2D-survey linear surface line-3
spread, mapping line of line-4
CMPs, b 3D-survey of
areal spread, mapping an
area of CMPs

CMPs CMPs
distance, x
time line-1 line-2 line-3 line-4
T

2D- shot record 3D- shot record

shot point, is usually referred to as the azimuth. centered about the CDP. The width and length of
Azimuth variation is considered as an advantage this search area are equal to the CDP spacing in
of the 3D data because with this feature the the two perpendicular directions of the survey
reflection seismic waves shall sample the rock station-grid. This search area is called the (bin)
medium in three-dimensional space and not and the traces included within it form the
restricted to one plane as the 2D data is provid- bin-gather. The process of sorting of traces into
ing. The concept is shown in Fig. 7.13. the appropriate bin is called bin-gridding or
binning. These concepts are shown in Fig. 7.15.
The bin is normally specied by certain
7.2.4 The CDP Bin and Bin Attributes properties or identication criteria. The bin can be
rectangular or square in shape, having a dened
In normal proling technique followed in 2D dimensions and surface area commonly referred
surveying, the spread moves along a uniformly to as the (bin size). Other parameters of the bin
spaced station-points, at equal move-up rate. In concerns its trace gather. These parameters, called
this way, the reflection points of each CDP the (Bin Attributes), include the trace offset, az-
trace-gather are all coinciding at the imuth, and fold of coverage. Normally, the bin
CDP. Sometimes, due to a shifted placement of a attributes are displayed in a special diagram
shot-point, one (or more than one) reflection called the spider diagram as shown in Fig. 7.16.
point may fall outside the CDP location. In CDP
sorting, a search distance (usually dened to be
equal to half of the CDP-spacing) is set in pro- 7.2.5 The Surface and Sub-surface
cessing. All traces found within the search dis- Coverage
tance of a CDP are included in the trace gather of
that CDP. In the normal swath shooting, the surface of the
The same principle is applied to the CDP-grid survey area becomes covered by two sets of
in case of 3D surveying, except that in 3D, we mutually perpendicular lines (the station lines).
have a (search-area) instead of the 2D These are the receiver lines and the shot-lines.
search-distance. The search area is normally The corresponding subsurface lines are the sub-
having a rectangular shape (or square shape) lines (inlines) and cross-lines which intersect at
7.2 Definitions and Basic Principles 151

Fig. 7.15 Denitions of (a)


a the search distance in 2D
surveying), and b search station-line
station point
area (Bin) in 3D surveying

CDP line
search distance

(b) station point


station-grid

CDP grid

search area
(BIN)

2 4
3 5 seismic surveying, the corresponding
end-product is a three-dimension space of
trace-1 offset azimuth stacked traces called the data volume or the
data box as it is sometimes called. The data
7 6 volume represents a three-dimensional function
in which the seismic amplitude varies with the
three coordinates; (x, y, t). The reflection-time
BIN CDP
(t) may be replaced by the depth dimension (z),
Fig. 7.16 Spider diagram showing the bin attributes where (z = tv/2) and v is velocity.
(offset, azimuth, and fold) of a bin having 7 traces in its The data volume (in digital form) consists of
gather (fold = 7) discrete volume-elements, called the (resolution
cells) dened as parallelepipeds of dimensions
bin centres. By denition, the bin is a subsurface (x, y, t), where (x) and (y) are the CMP
rectangular (or square) area having dimensions spacing in the inlines and cross-lines respec-
which are equal to half of the corresponding tively. The third dimension (t) is equal to the
surface receiver-station spacing (Fig. 7.17). sampling period of the stack traces. This volume
The subline and cross-line seismic sections element is considered to be the building brick of
are made up of sequence of stack traces, each of the data box, the ultimate product of the seismic
which is located at a bin centre. In this sense, 3D survey.
subline and cross-line seismic stack sections are The resolution cell is also called the (voxel),
considered as of subsurface locations and not in analogy to the term (pixel) used for the unit of
surface locations. a digital picture. The data volume is a subsurface
data-set, made up of voxels of dimensions (x,
y, t) as illustrated in Fig. 7.18.
7.2.6 The Seismic Data Volume With the help of special processing software,
it is possible to extract a variety of seismic sec-
The end-product of a 2D-data processing, is the tions from the data-volume. Normally, three
2D seismic stack section. In case of the 3D types of mutually perpendicular sections can be
152 7 3D Seismic Reflection Surveying

Fig. 7.17 Surface lines


(receiver and shot lines)
and sub-surface lines receiver
(sublines SL and cross-lines point receiver
XL). Bin dimensions are y line
half the corresponding
receiver spacing lines
x

sub-lines
(inlines)
(SL)
cross-lines
(XL) XL
CMP bin y
SL
receiver point x

Fig. 7.18 Denition of XL1 XL2 XL3 XL4


the resolution cell (the CD
voxel); the building brick SL4
of the seismic data volume SL3
y SL2
SL1
0
x
XL3 XL4
t

y SL2
x SL1
t
3D data box
(data volume) resolution cell
(voxel)

extracted. A vertical section may be obtained representation of the seismic amplitude values
along any subline, or along any crossline, and a which fall on the same reflection-time. In fact, it
horizontal section (called a time slice) can be is the locus of equal-time amplitudes existing in
obtained at any reflection-time. It is also possible the data volume. In color displays, the time-slice
to obtain a vertical section connecting points at plots are given in color-coded amplitude-values
arbitrary locations as for example, sections along with clear distinctions between peak- and
distances connecting several well locations. trough-values.
A section which is not a subline or a crossline is The time slice has a similar form as that pre-
normally referred to as a crooked, oblique or sented by the time contour map. The difference
diagonal section (Fig. 7.19). between the time slice and the structural contour
The horizontal section, called seiscrop sec- map is that in time slice, we display different
tion, or more often called (time slice), is a amplitude values at constant reflection time (time
7.2 Definitions and Basic Principles 153

Fig. 7.19 The 3D data


volume, with the four crossline
possible sections obtainable
from the data volume inline
(inline, cross-line, crooked
line, and time slice) crooked line

time slice

of the time slice). The structural contour map and data analyses are applied in the eld to
shows different time values of the same reflection determine the optimum parameters applied in the
surface. Briefly stated we say that time slice is data acquisition.
showing different amplitudes measured at one Here below, are the main aspects of the
time-slice, and in contour map, we display the data-acquisition requirements involved in 3D
same reflection event (belonging to a given seismic reflection surveying. Spread parameters
reflector surface) at different time values. Several and shooting procedure is most important aspect
reflectors may be represented in a given of the 3D eld acquisition activities. Many types
time-slice, but different time-values of the same of spread are in application, which can be dis-
given reflector, are shown in a given contour map. cussed under two main headings: marine and
Derivation of structural contour maps from land 3D spreads.
time-slice sections can, in principle, be obtained
by marking the amplitudes of the same seismic
reflector, on a sequence of time-slices, and pro- 7.3.1 Types of Marine 3D Spreads
jecting the picked values for that reflector on one
map. Conventionally, marine 3D surveying is con-
ducted by using spreads similar to those used in
land 2D seismic surveying. The boat and the
7.3 3D Field Data Acquisition towed-behind streamer is recording as it moves
along straight-line courses. The technique differs
In 2D seismic surveying, the shooting spread from the normal 2D surveying in that the survey
consists of a shot point and receiver points which shooting is conducted along closely spaced par-
are all arranged in one straight line (linear allel lines. Line spacing (typically, 100200 m),
spread). In the 3D surveying, however, these is kept constant throughout the survey area.
survey-points are distributed over an area (areal With the progress that took place in the nav-
spread). Essentially the eld work consists of a igation techniques and in navigation data pro-
set of procedures which implement certain geo- cessing, certain improvements were introduced
physical parameters. The eld parameters are on the applied spreads. In cases of calm seas, the
normally adopted through compromising survey boat and its trailing streamer move along
between the ideal technical requirements and the a straight line. In other cases, and because of
overall acquisition cost. Certain test-recording local sea cross-currents, the streamer may drift
154 7 3D Seismic Reflection Surveying

Fig. 7.20 The feathering


effect occurring in marine

surveying. The angle (h) is
the feathering angle

source
hydrophone receiver
recorded CMPs

away from the boat linear course. This is referred reflection points varies as shooting progresses.
to as the (feathering effect), and the angle Computations of the reflection points and CMP
between the drifted cable and the linear distributions are done in the processing stage
boat-track is called the feathering angle based on the gathered boat readings of the cable
(Fig. 7.20). geometry during shooting.
The shape of the drifted cable, for each shot, To increase survey efciency, multi-source
is determined from the readings of the special and multi-streamers have been used in the marine
compasses tted along the cable. Because of surveying. Examples of marine spreads which
changes of sea conditions, the cable shape and can be congured are: single source-single
the feathering angle change from shot to shot. streamer, dual source-single streamer, single
Consequently, distribution of the subsurface source-dual streamer, and dual source-triple
streamer (Fig. 7.21).
Spread movement during recording, is done
(a) normally along parallel straight lines (parallel
linear paths). Because of the long streamer-line
towed behind, the boat needs to turn with a rel-
(b)
atively large steering radius in order to record the
following line. A variation of this method is
(c) shooting along circular paths. The boat moves
along overlapping or spiral circular paths. The
advantages of the circular type of spread, is that
(d)
the survey is completed with no times lost due to
long-time turns of the boat needed in case of
source
hydrophone receiver straight-line surveying. These types of
spread-movements are shown in Fig. 7.22.
Fig. 7.21 Types of spreads used in marine 3D survey- It should be noted here that, deviation of the
ing. a Single source-single streamer, b dual source-single streamer course-line from the linear form (caused
streamer, c single source-dual streamer, d dual by the circular spread or by the feathering effect),
source-triple streamer

(a) (b) (c)

Fig. 7.22 Types of spread movements used in marine 3D surveying. a Parallel linear path, b Overlapping circular
path, c Spiral circular path
7.3 3D Field Data Acquisition 155

Fig. 7.23 Special types of (a) (b) (c)


land 3D spreads.
a X-Spread. b L-Spread.
c T-Spread.
d Square-Spread.
e Loop-Spread

(d) (e)

Source Receiver point CMP

gives certain amount of azimuthal variation. This The Rectangular Spread


makes the marine surveying to be approaching The most commonly applied shooting spread,
the proper 3D surveying. used in land 3D surveying, is the
rectangular-shaped spread, normally referred to
as the spread template. It consists of a number of
7.3.2 Types of Land 3D Spreads parallel receiver lines with a shot-line perpen-
dicular (or inclined) to the shot-line. The
The normal spread used in land 3D surveying shot-line is inclined with respect to the receiver
consists of source-lines perpendicular to receiver lines, usually at an angle of 45. The group of
lines. Examples of these special types of spread shots per shot-line is normally referred to as a
(not in common use at present) are X-spread, shot-salvo. A typical form of the
L-spread, T-spread, square-spread (called normally-applied rectangular spread (spread
Seis-square), and Seisloop (Fig. 7.23). template) is shown in Fig. 7.24.
In order to obtain multiple reflection points This type of spread is used in the
and increase of fold of coverage, several parallel land-surveying technique called (swath-shooting).
survey lines (source-lines and receiver-lines), are
used in the X-, L-, and T-spread. In the case of
the square spread, source-points and
receiver-points are distributed uniformly around
a square. If each survey point is occupied by a
source point and receiver point, the area dened
by the square becomes lled with CMP points of
high fold of coverage, distributed over a uniform
CMP-grid. If, however, the source points and
receiver points are laid out around an arbitrary
loop (loop-spread), the area enclosed by the loop
becomes lled with CMPs, but in an unorganized receiver point source point
distribution. Special software can be used in this Fig. 7.24 Spread template commonly used in Swath
case, to sort the created CMPs into a dened shooting. Source line is perpendicular to receiver lines.
bin-grid. Spread is made up of 7 receiver lines and 5-shot salvo
156 7 3D Seismic Reflection Surveying

overlapping spread positions

Spread
swathmovement along
first spread last spread

Fig. 7.25 Spread movement (shift-and-shoot movements) over the swath, with overlapping shooting spreads

7.3.3 The Swath Shooting Technique gives uniform bin fold and adequate offset and
azimuth coverage. On the other hand, swath
The swath is dened to be a strip of the survey shooting requires land with fully accessible space,
area of width equal to the width of the used and type of surface conditions which allow the
spread that moves along the dened strip in a appropriate freedom for the survey maneuvering.
roll-along movement similar to the A typical example, of shooting parameters
shift-and-shoot technique used in conventional applied in swath survey, is given here-below
2D proling surveying (Fig. 7.25). (Table 7.1).
In practice, when the template reaches the end
of the rst swath, the template is re-congured at
the near-by end of the adjacent swath, then car- 7.3.4 Types of Templates Used
rying out the shift-and-shoot process from end in Swath Shooting
to end. In this way all the rest of swaths are
sequentially covered. In general, swaths are not There are many types of templates that can be
laid down side by side, but laid down with used in swath shooting. A template may contain
overlap to obtain fold build-up (Fig. 7.26). one, or more than one, shot-lines which are
Greater fold coverage is obtained with bigger arranged to be in-centre, off-centre, or at ends of
overlap made between adjacent swaths. the template. The more common types are those
Swath shooting is a common acquisition where the shot-lines are perpendicular to the
technique implemented in todays land 3D sur- receiver lines. Another variation is to have the
veys. It has the advantages of being simple to shot-lines inclined with respect to the
congure, and efcient to execute in the eld. It receiver-lines (Fig. 7.27).

Fig. 7.26 Template


movements along the
sequence of ve
overlapping swaths
swath-5

swath-4

swath-3

swath-2

swath-1 template movement along the swath


template
7.3 3D Field Data Acquisition 157

Table 7.1 Typical values of operation parameters used in 3D swath shooting


1. Spread (template) conguration parameters

Spread shape Symmetrical split-spread


In-line offsets 252975 m
Receiver-point spacing 50
Receiver-line spacing 250
Receiver lines/spread 12
Active receivers/receiver-line 120
Source-point spacing 100
Source-line spacing 150
Number of shots/salvo 8
In-line spread roll (in number of receiver-points) 5
Cross-line roll (in number of shot-lines) 3
Nominal fold 24
2. Receiver array (geophone-group) conguration parameters
4.16 m

1.4 m
array centre 10.0 m

10.0
m
48.56 m
Geophone type/polarity SM-4/SEG standard
Natural frequency 10 Hz
Critical damping 70 %
String conguration 12 geophns (2 parallel  6 series)
Receiver array parallelogram
Geophones/geophone-group 36
Geophone spacing 4.16 m
Number of strings 3
Stagger between strings 1.4
String separation 10 m
Array length 48.56 m
Array width 20 m
(continued)
158 7 3D Seismic Reflection Surveying

Table 7.1 (continued)


3. Source array conguration parameters
12.5 m

5m
array centre

37.5 m
Source type Vibroseis System (vibrator)
Peak force 45,000 lb
Number of vibrators 4
Sweeps per VP 1
Sweep length 16 s
Sweep frequency 864 Hz
Sweep taper 500 ms at both ends
Source array Parallelogram 2  2
Vibrator inline stagger 12.5 m
Vibrator lateral spacing 5m
Array length 37.5 m
Array width 5m
4. Data recording parameters
Recording system I/O system-II
Channels recorded 1440 data, 4 auxiliary
Record length 6s
Sampling period 4 ms
Tape recording format SEGD
Recording gain 48 db
Diversity staking Active
Stacking/correlation Correlation after vertical stacking

(a) (b) (c)

(d) (e) (f)

Fig. 7.27 Types of templates used in swath shooting. a Centre symmetric split. b Asymmetric split. c Double off-end.
d Double within. e Centre multi shot-line. f Centre of inclined shot-line

More elaborate spread designs are based on shot-lines. One of such designs is what is called
grouping the receiver points in blocks with the Checkerboard spread. An example represent-
source points distributed regularly over parallel ing this type of spread is shown in Fig. 7.28.
7.3 3D Field Data Acquisition 159

area. Examples of inaccessible areas are farms,


built-up blocks, lakes, or isolated rugged moun-
tains. 3D surveying of the totally inaccessible
area can be carried out using a technique called
Loop Shooting. This technique involves laying
out source-points and receiver-points around the
inaccessible area, forming a loop that encloses
the project area (Fig. 7.29).
While sequentially shooting the source points,
the receiver points, occupied by the geophone
groups, are xed in place. When the surveying
process is completed, the enclosed area becomes
lled with CMPs without having to move any of
the survey equipments inside the area. Special
processing software can be used to sort the CMPs
into a dened bin-grid.
Loop shooting furnishes a solution to the prob-
Fig. 7.28 A checkerboard spread consisting of one
8-source shot-line with 16-receiver blocks. Each block lem of area inaccessibility, but not without a cost.
includes 48 geophone groups (48 live seismic channels) To start with, it gives non-uniform fold and offset
distribution among bin gathers. Further, shallow
The template, in this checkerboard example, reflectors may be missed in case of too large
consists of one 8-source shot-line with survey-areas. However, with proper design of the
16-receiver rectangular blocks, each of which loop-route and using the appropriate processing
includes 48 geophone groups (48 live seismic software, this method can give fully 3D-surveyed
channels per block). area, especially when the area is not too large and
the target reflectors are not too shallow.

7.3.5 The Loop Shooting Technique


7.3.6 The 3D Survey Design
There are cases where the project area is totally
inaccessible (no survey equipment can enter the The principal objective of any 3D survey is to get
area) or partially accessible, where equipment a fully migrated seismic data volume of the target
can be moved along restricted routes crossing the subsurface geological structure. Quality of the

(b)
(a)

Source point Receiver point CMP

Fig. 7.29 Loop-Spread. a Survey points (source and receiver points) laid down around the irregular inaccessible area.
b Survey points laid down on the perimeter of a rectangle (or square) enclosing the survey area
160 7 3D Seismic Reflection Surveying

nal results is controlled by the procedures as Surface-area Parameters (area boundary,


well as the parameters applied in the acquisition full-fold zone, migration-fringe width).
and processing of the recorded data. In order to Sub-Surface Parameters (bin dimensions,
get clear reflection signals, optimum acquisition offset-range, azimuth range, fold of coverage).
parameters must be applied in conducting the
In order to conduct the computations and data
survey. Determination of the procedures and the
analysis required by the survey design, a number
acquisition parameters is involved in a special
of basic geological and geophysical parameters
group of eld activities, collectively known as
must be available. In particular we need to have
the (3D survey design).
information about the general geological nature
Pre-denition of depth-range of target reflec-
of the area, dip variation (magnitude and azi-
tors and considering the limitations of allocated
muth), velocity-eld, frequency-range, depth and
cost and time, serve as guide lines in the design
extensions of target reflectors. These aspects will
process. Acquisition eld parameters which need
serve as input to the survey-design analyses
to be optimized in the survey design include the
which give the optimum acquisition parameters
following actions:
to be applied in the survey execution.
Source Parameters (charge size, charge depth, Acquisition parameters covered in a typical
vibroseis parameters, shot pattern). survey-design process can be summarized as
Receiver Parameters (geophone-array shape, follows:
number of geophones per group, geophone
(i) Survey Area and Migration Fringe
spacing, near offset, far offset).
(ii) Template Parameters
Spread Template (template geometrical
(iii) Swath Parameters
shape, number of active channels per receiver
(iv) Bin Dimensions and Bin-Attributes
line, number of receiver lines per template,
(v) Sampling Period and Record Length.
number of source points per shot-line, num-
ber of shot-lines per template, spread
(i) Survey Area and Migration Fringe
move-up distance along swath).
A 3D seismic survey is designed in such a way
Swath Parameters (swath direction, roll-aside
that the target data-volume (areal extent and
overlap).
recording time-length) is obtained with full
Recording Parameters (sampling period,
migration. An additional full-fold strip
record length).

Fig. 7.30 Denition of surface area


the full-migration area and migration
the migration fringe in
subsurface area
fringe
respect to the full-fold, full fold area
subsurface, and surface full-migration area
areas

migration migration
aperture aperture
7.3 3D Field Data Acquisition 161

Fig. 7.31 Computation of


maximum offset (Xmax) and X min
minimum offset (Xmin) from
the geometry of the
template
X max

source point receiver point

surrounding the target migrated area is required pre-dened minimum to maximum offset values.
to insure full migration of the target area. The As a rule of thumb, an offset should be roughly
width of this strip (called the migration fringe) is equal to the depth of the target reflector. Thus, by
equal to the migration aperture which is dened dening the depths of target reflectors, it is possible
to be the horizontal distance a reflection event is to determine a range of offsets to suit both of the
moved by a migration process. The width of the targeted shallow reflectors and deep reflectors.
migration fringe is function of the dip measured For a given bin-size, geophone group spacing
in the direction of the fringe width. The migra- becomes known, and both total active channels per
tion fringe is shown in Fig. 7.30. receiver-line and total number of receiver-lines per
The survey area should be large enough to template can be decided, based on the maximum
include the fully migrated zone, migration fringe, offset required. Maximum offset (Xmax) and mini-
tail-fold zone, and the required surface- mum offset (Xmin) can be calculated for the largest
coverage zone. For economic reasons, and when and smallest source-to-receiver points in the
the S/N ratio is sufciently high, the migration designed template (Fig. 7.31).
fringe may be extended in such a way as to include
(iii) The Swath Parameters
part of the tail zone. This step can result in
The main parameters of the swath are shooting
reduction of the total survey area to the area which
direction (survey orientation) and roll-aside
is sufcient for achieving the targeted
overlap. Swath length and swath width are con-
fully-migrated data volume.
trolled by the survey-area boundaries and tem-
Estimation of the fringe area is considered to
plate width respectively. Based on the
be an important issue in 3D surveys, since it is
fold-computation formula, the fold is function
directly connected with the overall survey cost.
of both of the template movement-overlap and
The percentage cost-increase due to addition of
swath role-aside overlap. Hence the overlap
migration fringe depends on the size of
(measured in terms of number of receiver-stations
survey-area as well as on the fringe width used.
for the template movement and in terms of
Cost saving, as we mentioned above, can be
receiver-lines for swath movement) can be com-
effected by allowing the fringe to be extended
puted once the fold is dened (Fig. 7.32).
into part of the tail-fold zone. According to
The other important parameter related to the
(Sheriff and Geldart 1995, p. 452), the fringe area
swath is the shooting direction, or survey orien-
should be equal to the migration aperture plus the
tation. When there is a distinct dip-trend in the
radius of the rst Fresnel zone.
area, receiver lines are oriented in the direction of
(ii) Template Parameters dominant dip. In practice, this constraint is
The template is made up of a number of relaxed for two reasons: rst, dip value and dip
receiver-lines, shot-lines, number of active chan- direction are generally not constant for the same
nels per receiver-line, and number of source-points area, and second, the dip move-out effect can be
per shot line. A template must be so designed that it taken care of, by dip move-out (DMO) correction
will provide a range of offsets covering a or by pre-stack migration in the processing stage.
162 7 3D Seismic Reflection Surveying

Fig. 7.32 Fold build up swath


by a overlapping spreads (b)
roll-aside
rolling-along the swath and
by b overlapping swaths direction
rolling-aside across the area

(a) spread roll-along direction

(iv) Bin Dimensions and Bin Attributes formula (Liner 2004, p. 272; Sheriff and Geldart
By denition, the bin is a subsurface rectangular 1995, p. 452):
(or square) area having length and width
Dx  v=4f sinh
dimensions. These dimensions are equal to half
of the corresponding surface receiver-station
where (v = f k) is velocity, (f) is the highest
spacing. In processing, seismic traces are sorted
frequency-component, and (h) is the angle of dip
into bin gathers. This means that the bin will
component in the x-direction.
contain a number of traces (equal to the fold of
Derivation of the formula can be done with
coverage) of different offsets, and different az-
the help of Fig. 7.33.
imuths. The centre of the bin marks the CMP at
This formula expresses the relation between
which the stack-trace of the bin-gather is located.
the CMP trace spacing (x) and the three vari-
Thus, the bin area (normally referred to as the bin
ables (v, f, h). It has shown that (x), which is
size) should be small enough to achieve appro-
representing the bin dimension in the x-direction,
priate resolution but not too small to create
is directly proportional to velocity (v) and
empty or low-fold bins. An optimum bin size
inversely proportional to wave frequency (f) and
must, therefore, be determined.
reflector dip (h).
The theoretical basis used for computing the
Typical bin dimensions used in land 3D surveys
optimum bin size is the same as that used for
are (12.525) m in the inline (sub-line) direction
avoiding spatial aliasing in case of 2D common
and (2550) m in the cross-line direction. These
mid-point (CMP) spacing computations. Thus, in
ranges are corresponding to (2550) m and
order to avoid spatial aliasing, at least two sur-
(50100) m of surface sampling intervals (receiver
face samples are required to be present for each
spacing). However, when square and small enough
apparent wavelength. For a given reflector dip,
bins are used, no spatial aliasing occurs. Small bins
the apparent wavelength should be as small as
mean ne subsurface sampling which is necessary
possible in comparison with the CMP spacing.
for increasing structural resolving power.c
From the geometry of the normal incidence
reflection from a dipping reflector and using Bin Attributes
(x = k/2 and t = s/2), a quantitative evalua- The group of traces included in a bin gather, are
tion is obtained for the maximum spatial sam- traces of different offsets recorded by receivers
pling interval (x), expressed by the following positioned at different azimuths. The number of
7.3 3D Field Data Acquisition 163

Fig. 7.33 Theoretical /2 /2


basis for computing the
maximum spatial sampling
interval (x), which is the
CMP trace interval x x

wave-front at
(t v/2) time, t+t

wave-front at
time, t
Sin =(t v/2) / x dipping
t = /2, = 1/f reflector
hence, x = v/4fsin

traces in the bin-gather represents the fold of Nyquist frequency is 250 Hz. which is expected
coverage for that bin. The parameters (offset, to be far higher than the highest frequency of
azimuth, and fold) form the bin attributes. recorded reflection signals. For economic rea-
In the design process, it is aimed at sons, the recorded data is often re-sampled to
attribute-values that cover as wide coverage-range 4 ms before being inputted to processing. To
as possible, considering the economic constraints. make sure that no aliasing occurs, an anti-alias
Bin attributes are normally displayed in a special lter is applied in the acquisition recording stage
diagram (called spider diagram) in which offsets prior to AD conversion and it is also applied
are drawn as straight lines radiating from the bin prior to any re-sampling process carried out
centre. The line length is proportional to offset during processing work.
value and line bearing represents the azimuth. The sampling period, together with the bin
Number of lines is equal to the fold of coverage dimensions, set the three dimensions of the res-
(Fig. 7.34). olution cell (the voxel) which has the general
In addition to the desired large coverage-ranges form of a parallelepiped or a cube. Its volume is
of attributes, the bin should be small and square to dened by the two horizontal dimensions (x
insure equally-ne sampling in both of the and y), and the vertical dimension (vt/2),
sub-line (inline) and cross-line-directions. where (x and y) are equal to the bin dimen-
sions in the sub-line and cross-line directions,
(v) Sampling Period and Record Length
(t) is the sampling period, and (v) the seismic
The sampling period normally used in the seis-
velocity (Fig. 7.35).
mic recording is 2 ms. With this value, the

XL1 XL2
(a) (b) (c)
SL2
y
x SL1
t

Fig. 7.34 Bin attributes, a poor azimuth coverage, CDP


fold-5, b fair azimuth coverage, fold-9, c good azimuth
coverage, fold-12 Fig. 7.35 Denition of the resolution cell (voxel)
164 7 3D Seismic Reflection Surveying

Size of the voxel is considered to be as mea- or,


sure of the 3D resolution power, where the spatial Dxm vTm =2 tan hm
resolution is governed by the dimensions (x and
y) and the depth resolution is governed by the giving,
sampling period (t). In a 3D data box where the
bin dimensions, (x = 25 m and y = 50 m) and Ts Tm sec hm
sampling period (t = 2 ms), the voxel volume is
(25  50  3000  0.002/2) m3, assuming the This formula shows that the migration aper-
velocity (v) to be 3000 m/s. Naturally, the smaller ture (xm) is directly dependant on the three
the voxel the better is the resolution power in both factors; average velocity (v), dip (hm), and
horizontal and vertical directions. two-way recorded reflection-time (Ts), which is
related to the true reflection time (Tm) by the
Record Length
relation (Ts = Tm sec hm).
From our discussion of the design of the survey
If the dip is 45, for example, migration of an
area, it is noted that the subsurface target area must
event at true time (Tm) needs to include data
be extended by an additional strip (the migration
down to (Ts = Tm sec 45 = Tm 2 = 1.4 Tm).
fringe) in order to insure complete migration. For
This means that an extra 40 % of the true time of
the same reason (getting complete migration), all
the dipping segment needs to be included in the
the seismic wave-eld originating from the sub-
migration process in order to get complete
surface target volume should be included in the
migration. Thus recording should be down to a
migration processing. This requires that a mini-
greater time than that of the deepest target (by
mum recording time which is of extent that can
extra time of about 40 %).
include all diffraction energy of the deeper parts of
the tails of the diffraction hyperbolae.
From the geometry of the migration of dip-
7.4 Processing of 3D Data
ping reflector, the aperture radius (xm) is given
by Fig. 7.36:
Seismic data processing, in all 2D and 3D sur-
Dxm vTs=2 sin hm veys, aims at enhancing the reflection signal and
getting accurate images of the subsurface

Fig. 7.36 Denition of


the migration aperture
x
(xm) and the required
minimum recording time
(Ts)
vTm/2
vTs/2

m
apex

circular m
wave-front
diffraction
s hyperbola
7.4 Processing of 3D Data 165

Fig. 7.37 A general


processing sequence for 3D Field Data
data processing. SHOT-G,
BIN-G, and STK-SEGY Seismic Data Statistical Data
represent optional
shot-gather, bin-gather, and
migrated stack data volume 1. DATA REORGANIZATION
(in SEG-Y format) - Data reformatting & loading
respectively - Demultiplexing
- Geometry & Statics computations
- Trace-header assignment
- Re-sampling
SHOT-G.
2. PRE-STACK PROCESSING
- Trace editing
- Coherentnoise attenuation
- Randomnoise attenuation
- Spherical Divergence Compensation
- Surf. Consist. Amplitude Compensation
- Field-statics correction
- Deconvolution
- Bin-gather sorting

BIN-G.
3. PARAMETER OPTIMIZATION
- Velocity analysis
- Residual-static analysis
- Deconvolution analysis
- Filter analysis

Residual statics correction

NMO-DMO corrctn. Migration


First arrival muting (PRSTM)
Stacking
Trace interpolation
NMO-correction
Migration First arrival muting
(POSTM) Stacking

4. POST-STACK PROCESSING
- Zero-phase deconvolution
- S/N Enhancement (f-xy deconvolution)
- Band-pass filtering
- Equalization
- QC displaying

STK-SEGY
166 7 3D Seismic Reflection Surveying

geological structure. For processing-steps may be grouped into four major


stratigraphic-exploration purposes, however, divisions. These are:
data-processing is directed towards preservation
data re-organization,
of the amplitude changes caused by the physical
pre-stack processing,
properties as well as structural changes of the
parameter optimization,
subsurface rock layers. Whether it is obtained
post-stack processing.
from 2D or 3D surveying, the recorded seismic
reflection data is subjected to a sequence of A basic processing sequence including this
data-analyses (data processing steps) which give four-stage processing-procedure is summarized
the nal stacked 2D seismic sections or 3D by the flowchart shown in Fig. 7.37. This gure
seismic data volume. shows a sequence of the four groups of pro-
cessing steps; data re-organization, pre-stack
processing, parameter optimization and
7.4.1 The Field Data Processing post-stack processing. The migration process is
shown in two alternative positions. One option
Field crews are usually equipped with processing (the more common practice in nowadays pro-
facilities (eld processing system) used to pro- cessing) is the pre-stack time migration
cess data for some eld parameters determination (PRSTM) and the other option is the application
and for QC purposes. Field processing is mainly of post-stack time migration (POSTM). In this
done to analyze noise-test data and to determine sense, migration is considered either part of the
source parameters (charge weight, charge depth, pre-processing stage (PRSTM) or part of the
shot pattern, sweep parameters), in addition to post-processing stage (POSTM).
the preliminary eld processing. The eld crew
may shoot a short 2D test-line, which needs to be
processed onsite rather than in the processing 7.4.3 3D Data Displays
centre. In order to save time, these processing
activities are usually done by the crew processing The ultimate end-result of the processing
geophysicist using the crews processing facili- sequence is a data-volume, which is also, called
ties, rather than sending the data to the far-away data box (Fig. 7.38).
processing centre. The data volume consisting of an amplitude
value at the point (x, y, t), where (x) and (y), are
the bin-centre coordinates in the in-line and
7.4.2 The Final Data Processing cross-line directions respectively. The coordi-
nates (t) is the time value of the trace sample.
Once the shooting parameters are nalized, nor- Three types of displays are usually extracted
mal production shooting is started and shot from the produced data volume: an in-line,
records (recorded on magnetic tapes) are cross-line vertical-sections, and horizontal sec-
shipped-out to the processing centre to be pro- tions (time-slices). At each sample-time, a
cessed. As it is done with 2D data, processing of time-slice can be displayed. Examples of vertical
3D data is conducted through a sequence of and horizontal sections are shown in Figs. 7.39
processing steps. The sequence of 3D and 7.40.
7.4 Processing of 3D Data 167

Fig. 7.38 The 3D data volume obtained from processing of a 3D data-set of an area of (10 by 20) km2

in-line (sub-line) cross-line

Fig. 7.39 Vertical sections (in-line and cross-line) obtained from a 3D data volume
168 7 3D Seismic Reflection Surveying

time-slice, color-coded time-slice, in black & white

Fig. 7.40 Time slices obtained from the 3D data volume


The Seismic Reflection Signal
8

A seismic eld is created as a result of a sudden nding a way to apply statistical communication
mechanical impact that occurs at a point inside a theory to the seismic problem (Enders Robinson
medium. The created packet of energy (called the 1983, p. 221).
seismic energy pulse) spreads out from the The GAG research activities have resulted in
source zone in all directions. It travels through furnishing a way for applying the principles of
the medium in such a way that energy moves the communication theory in the analysis of
through the particles of the medium without seismic wavelets. In essence, the seismic wavelet
creating any permanent changes to the medium. is considered as a propagating signal which can
Propagation of the generated seismic energy in be processed using principles of the communi-
this way is expressible in terms of a mathematical cation theory in the same way as processing of
equation, the (wave-motion equation). the electromagnetic signal.
In a seismic reflection exploration that uses The GAG scientic ndings in the late 1950s
impulsive energy-sources, the generated seismic have set the cornerstone of modern digital pro-
pulse is generally made up of an oscillating cessing of the seismic signal (the seismic
energy level that diminishes with time and with wavelet).
travel distance, forming what is normally called
the (seismic wavelet). This travelling pulse (the
wavelet) can be recorded by placing a seismic 8.1 Definition of the Seismic Signal
detector (seismometer) in, or on the surface of,
the medium. The seismic wavelet detected near In general, the travelling seismic wavelet (de-
the source location is normally referred to as the tected directly, or after being reflected or refrac-
source wavelet (or source signature), and that ted) is considered to be a travelling signal that
detected after being reflected from an interface, it bears useful information. The seismic reflection
is called the reflection wavelet. signal, normally representing the particle-
In 1952, Massachusetts Institute of Technol- vibration velocity, is a function of space and
ogy, MIT (USA) set up and sponsored a research time. The value of the function at any time is
group, called the Geophysical Analysis Group dened to be the amplitude (or energy) of the
(GAG) which was in 1953 taken over by a signal. Usually, the seismic signal is presented as
consortium of oil and geophysical companies. a mathematical function that expresses amplitude
This group of scientists had the responsibility of variation with time (Fig. 8.1).

Springer International Publishing Switzerland 2017 169


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_8
170 8 The Seismic Reflection Signal

and signal energy changes (due to physical


amplitude
properties of the medium).
Types of changes can be sorted into three groups
f(t)
according to the zone in the reflection travel-path
time, t from source to the receiver. The main types of
changing factors are presented in Table 8.1:
It is important to note that each of the factors
mentioned in this table, has its own
modication-effect (imposed signature) on the
Fig. 8.1 Seismic signal is represented by plotting the
signal amplitude f(t), as function of time, t travelling seismic signal and what is received, at
the end of the source-to-receiver journey is the
net resultant of all the effects encountered during
8.1.1 The Seismic Signal that journey. These changes (travel-time and
Parameters waveform changes) have their own individual
imprints (signatures) imposed on the travelling
Treated as a mathematical function, the seismic wavelet. These effects form useful messages
reflection signal has a number of parameters (information) which can be interpreted to give
which dene its overall characteristics. These the subsurface structural and stratigraphic chan-
parameters, which are generally changing with ges. The travelling seismic signal (the seismic
travel-time, are: wavelet) with the modications it experiences
Amplitude (signal strength) during its source-to-receiver journey form the
Frequency content basis of its use as an oil exploration tool.
Energy content
Propagation velocity
Onset time 8.1.2 The Seismic Signal Spectrum
Duration (signal length)
If the medium, traversed by the seismic signal, By using Fourier Transform, it is possible to
is perfectly elastic and homogeneous, the signal transform the mathematical function of any sig-
will propagate along straight ray-paths with nal (as the seismic signal) from its original
constant velocity and with no changes in its time-domain function f(t) to the frequency
waveform parameters, apart from the domain F(x) with no loss of information. In this
wave-spreading effect. In reality, the earth is far process the original time domain signal was
from being homogeneous, as it is made up of transformed from amplitude variation with time
rock-layers having different physical properties to amplitude variation with frequency. To clarify
and different geometrical shapes. Geological the concept let us consider (Fig. 8.2) the two sine
complexities in both of the structural and strati- waves of different frequencies (x1 and x2), and
graphic nature, incur changes on the seismic their respective line-spectra drawn at points x1
wavelets shape, velocity, and motion direction. and x2 on the frequency axis of the spectrum
The changes are of two types; changes of function, F(x).
signal-path geometry (due to interfaces effects)

Table 8.1 Changing factors of the seismic reflection signal during its sourceto receiver travel-path
Changes in source zone Changes in ray-path zone Changes in receiver zone
Source type of energy source Inelastic attenuation Receiver detection response
Source coupling Geometrical spreading Detector coupling
Near-source geology Reflection coefcient Near-receiver geology
Source-generated noise Wave conversion Noises and interferences
Noises and interferences
8.1 Definition of the Seismic Signal 171

f(t)
F()
t
1

g(t)
G()
t
2

f(t) + g(t) F() + G()

t 1 2

Fig. 8.2 Concept of the frequency spectrum. The mono-frequency functions, f(t) and g(t) have one line spectrum each.
Their sum, f(t) + g(t), have two-line spectrum

time domain function

A0 time (t)

Sine functions, A(t) amplitude spectrum


A1

A2
A( )
a0 a1 a2 a3 a4 a5
A3
1 2 3 4 5
A4
phase spectrum
()
A5

Fig. 8.3 Pictorial representation of spectrum analysis. The signal (at top of gure) is Fourier-transformed into its
frequency components shown as amplitude and phase spectra, A(x) and (x)

This simplied presentation represents a syn- present in the analyzed signal. Two spectral parts
thesis process by summing two sine functions. are necessary to completely dene the complete
The reverse of this process is the spectrum anal- signal characteristics. These are the frequency
ysis which involves (using Fourier transform) spectrum and the phase spectrum. The two con-
determination of all of the frequency components cepts are schematically shown in Fig. 8.3.
of a given signal. The complete analysis is to get The frequency range of the seismic reflection
amplitudes and phase shifts of all frequencies signal is typically (2060) Hz.
172 8 The Seismic Reflection Signal

Table 8.2 Differences Signal parameters Seismic signal Radar signal


between the seismic signal
and the radar Type of ergy Mechanical energy Electromagnetic energy
electromagnetic signal Type of medium Elastic medium Empty space
Pulse length 1030 s (Vibroseis) 106 s
Velocity 15 km/s 3  105 km/s
Frequency 2060 Hz 1010 Hz
Wavelength 20500 m 3 cm

8.1.3 Comparison with the Radar conditions. The manner and severity of these
Signal changes depend largely on:
Geometrical shape of the traversed layers
Radar signal and seismic signal are similar, in that (structural changes)
they both provide the source-to-receiver distance Physical properties of the layers (stratigraphic
of a far-away object, using the wave-reflection changes)
phenomenon. The source signal used in the case of Fluid content like water, oil, and gas (reser-
radar is typically a short pulse (of a wide voir conditions)
frequency-band), a chirp signal (linearly-changing The seismic wavelet is therefore considered as
frequency signal), or a frequency-modulated a message-bearing record of the changes which
(FM) signal. The radar function is based on two are coded in the wavelet spectrum-parameters
main types of measurements; the travel-time and (amplitude, frequency, and phase). In the
Doppler-effect measurements. The Doppler Effect time-domain language we say that we have
(variation of frequency with target motion) is waveform-deformation whereas in the frequency
made use of, in velocity measurement. The com- domain this is described as spectral-deformation.
mon feature between radar operation and seismic Compared with the source function, and due
reflection technique is that both of them use the to these affecting factors, the detected reflection
reflection travel-time measurement. Naturally, signal becomes lower amplitude, lower fre-
Doppler Effect has no application in seismic quency, extended in shape, noisy, and
reflection operation. time-shifted with respect to its initial origin time.
The main differences between the seismic Mathematically, these changes can be classied
signal and the electromagnetic signal used by into the following four types of processes:
radar are summarized in Table 8.2. superposition, scaling, ltering, and time
This table shows that the parameters of the shifting.
seismic signal are dispersed over a wide range of (i) Superposition
values, compared with those of the radar signal in Interaction of the reflection seismic signal
which the parameter has only one value. This is with noises and other interfering events
because the radar signal is travelling through a (such as coherent and incoherent noise,
medium which is far more homogeneous than refractions and diffractions) is a simple
that traversed by the seismic signal. superposition process. The received signal
is resulting from additions of all of these
interferences, bringing about distortions to
8.1.4 The Seismic Reflection-Signal the wavelet geometrical shape.
Changes (ii) Scaling
This is an arithmetic multiplication pro-
The seismic reflection wavelet is, in fact, the cess, where the reflection wavelet is mul-
source pulse after being modied and distorted as tiplied by a constant scaling factor, similar
a result of the interactions with the travel-path to the effect of gain-application. The effect
8.1 Definition of the Seismic Signal 173

Fig. 8.4 Summary of the - high amplitude (i) superposition - lower amplitude
physical changes of the - high frequency content (ii) scaling - lower frequency content
reflection signal as it travels - sharp shape (iii) filtering - extended
from the source point (S) to - with no noise (iv) time shifting, to - noisy
the detector point (R), via - Onset-time, t=0
- time Shifted, t=t0
the reflection point (RP) S R
source impulse reflection wavelet
V1
s(t) r(t)
t V2 t
0 0 t0
V3

V4

RP

does not deform the signal but only detector offset from the source.
change its size. Examples of this type of Time-shifting becomes a deformation
effect are the effect of the geometrical factor when reflectors are so close to each
spreading and the reflection coefcient other that wavelet-arrivals overlap causing
which are independent of frequency. This distortions to the wavelets and loss of the
means that they incur no wavelet shape resolution of the reflection events. Phase
deformation. A scaling effect can, under shifts are the time shifts of the individual
certain conditions, reverse the wavelet frequency components which, in general,
polarity. This happens when the scalar is take place in frequency-ltering actions.
of negative value, as in negative reflection These changes are summarized in Fig. 8.4.
coefcient.
(iii) Filtering
Frequency ltering can be considered as a 8.2 The Seismic Trace
frequency selective scaling. A typical
example of this type of effect is the earth A seismic pulse incident at an interface is partly
high-cut lter, which is operating on the reflected and partly transmitted into the following
seismic wavelet attenuation throughout its layer. In a multi-layer medium this is repeated at
travel-path. Any effect that is each interface present in the way of the advanc-
frequency-dependent (as the earth lter- ing wave. A detector placed on the surface
ing) incurs shape-deformation to the receives sequentially, the reflected wavelets, at
propagating seismic wavelet. The defor- time-intervals depending on the depths of the
mation severity depends on the type of reflectors. The source pulse reaches the detector
lithology and fluid-content of the geolog- after being affected by the various types of
ical formations traversed by the seismic modications (as discussed above) including the
signal. Being an analogue-type, the earth process of reflection in which amplitude and
ltering effect introduces phase shift to the polarity are governed by the reflection coefcient
frequency components of the seismic of the reflector. From each reflector a wavelet
signal. representing the source wavelet, scaled by the
(iv) Time-Shifting reflection coefcient, will arrive at the detection
The source impulse takes time to reach the point. Thus, a series of such wavelets, each of
detector. This is the total reflection which is shifted in time with respect to the pre-
travel-time which is depending upon the ceding wavelet, are superimposed on each other
depth of the reflector as well as on the to form a record which is the (seismic trace).
174 8 The Seismic Reflection Signal

reflection seismogram (seismic trace)


geological acoustic reflection
section time
impedance coefficient
(-) (+)

Reflector-1
1

Reflector-2
2

Reflector-3
3

Reflector-4
4
Reflector-5
5

Fig. 8.5 Sketch of the convolutional model of the seismic trace (Alsadi 1980, p. 192)

8.2.1 The Convolutional Model where,


of the Seismic Trace
f(t): recorded seismic traceSeismic trace
s(t): seismic source wavelet
The process, which involves multiplication of the
p(t): earth ltering effect
source wavelet by the reflection coefcient, time
r(t): earth reflectivity function (series of the
shifting and superposition, is similar to the
reflection coefcients)
mathematical convolution process taking place
b(t): detector and recording equipment
between the source function, in this case, and the
response
reflection-coefcient series. This is the con-
g(t): geometrical spreading effect (scaling
struction mechanism model which is accepted by
effect)
geophysicists for the formation of the seismic
n(t): noises
trace. This model (called the convolutional
model) is applied in working out synthetic seis-
It is to be noted here that convolution pro-
mograms (Fig. 8.5).
cess is represented by (*), whereas the effect of
The concept that the seismic signal can be
geometrical spreading, g(t), is purely scaling
represented as the convolution of the source
effect, and noises, n(t), has an additive effect
wavelet with the earth reflectivity-series was
(Fig. 8.6).
developed by Robinson (1954). This approach
Typically, a seismic trace recorded in a seis-
(based on the communication theory) can be
mic reflection survey, is 6-s long, digitally
represented by the following mathematical form
recorded at 2 ms sampling period (Fig. 8.7).
(called the seismic-trace convolutional-model
equation):

f(t) = s(t) * p(t) * r(t) * b(t) . g(t) + n(t)


8.2 The Seismic Trace 175

Fig. 8.6 Elements


entering in the seismic trace f(t) = s(t) * p(t) * r(t) * b(t) . g(t) + n(t)
convolutional model

The convolutional model of the seismic trace

Seismic trace

Source function

Earth filter

Reflectivity

Instrument. Resp.

Geom. Spreading.

Noise
Fig. 8.7 The seismic
reflection trace. a Pictorial (a) (b)
representation. b Actual amplitude amplitude
seismic trace from a
0
dynamite source-energy

1
reflection time

2
in
3
seconds

8.2.2 The Synthetic Seismogram The earth layering model is normally obtained
from a drilled exploration-well. The well logging
A direct application of the concept of the con- data furnish both of the velocity (from the sonic
volutional model is in constructing an articial log) and density (from density log). From these
seismic trace, commonly known as the (synthetic two logs, the acoustic impedance of each layer is
seismogram). The basic requirements for this calculated, and hence the reflection coefcients
process are the source wavelet and the reflectivity of the layer boundaries are determined. This is
series, which in turn, require the distribution with the reflectivity series which will be in the form of
depth of the layers acoustic impedances. values, ranging between (1) and (+1),
176 8 The Seismic Reflection Signal

Fig. 8.8 Pictorial


representation of the
principle of computing
synthetic seismograms

(x) (*)

velocity density reflectivity wavelet Synth. seismogram

Fig. 8.9 Application of


the synthetic seismogram,
which is duplicated and
superimposed on the
seismic stack section at the
well location

representing the reflection coefcients of all of with the real geological section. It is normally
the interfaces (layer-boundaries) penetrated by displayed (using several duplicated synthetic
the well. Convolution of the source wavelet traces) superimposed on the seismic stack section
(assumed or obtained from a near-by shot-record) at the well-location. Main purpose is to attach the
with the reflectivity series, gives the synthetic stratigraphic identities for the reflection events
seismogram. This computation method expresses diagnosed on the seismic stack section (Fig. 8.9).
the convolutional model according to which the
real seismic trace is assumed to have taken place
in real time seismic reflection recording. 8.2.3 The Digital Seismic Trace
In display of the computed synthetic seismo-
gram, the time scale must be doubled in order to The seismic trace received at the detection point
match the seismic section in which the time is a is represented as a continuous amplitude varia-
two-way vertical time. A schematic representa- tion with recording time. At this point it is an
tion of the process is shown by Fig. 8.8. analogue function expressing the seismic ampli-
Interpreters are usually provided with syn- tude as function of time. The recording-system
thetic seismograms both with and without multi- transforms the amplitude-versus-time function
ples to help in sorting out the multiple-reflections, from its analogue form to digital form, and gets it
if any. stored on the magnetic tape as a digital function
The synthetic seismogram is now considered (sequence of sample-values at regular time
as an indispensable tool in the hand of the intervals, t), where the sampling period (t) is
interpreters to tie the seismic reflection images normally chosen to be 2 ms (Fig. 8.10).
8.2 The Seismic Trace 177

Fig. 8.10 The digital


seismic trace, made up of a seismic trace analogue mode digital mode
series of sample values
recorded at regular time - 0 + - 0 +
interval, which is the
sampling period (t)

TWT

With the digital mode, the seismic trace stor- 8.3.1 Definition of the Seismic
age and processing become feasible to carry out Wavelet
by electronic digital computation systems.
A wavelet is dened as a transient signal with a
denite time origin. It is characterized by two
8.3 The Wavelet Concept properties; it has a denite onset time and it has a
nite energy (Robinson 1983, p. 128). These two
The seismic pulse created by a mechanical shock characters implies that the wavelet is one-sided
(seismic source) is a wavelet that is transmitted entity (wavelet values are zero before onset time),
through the earth medium and detected by the and it is transient with its energy diminishing after
receiver geophone-group and then recorded by onset time. Ricker wavelet (named after Norman
the recording system. At the time it is recorded, Ricker 18971980) is a special type of a
the wavelet is no longer a short pulse (source theoretically-computed zero-phase wavelet com-
impulse) as it was created at the source location, monly used in seismic modeling studies. In par-
but an extended wavelet made-up of few cycles, ticular, Ricker wavelet is used in the seismogram
weakened pulse having energy diminishing to synthesis, by convolving it with the reflectivity
zero level within a short time (tens of millisec- function, to produce the now-accepted convolu-
onds) (Fig. 8.11). tional seismic trace model (Robinson 1983, p. 224).

Fig. 8.11 The seismic


wavelet function of
travel-time, f(t). Its value (a) (b)
represents the particle
vibration velocity in land f(t) f(t)
surveys, and hydrostatic
pressure in marine surveys.
a is the source wavelet and
b is the recorded weakened t
wavelet t
178 8 The Seismic Reflection Signal

Fig. 8.12 Different types


of seismic wavelets.
a high-frequency (short
wavelength, kS) as that (a) (b)
used in sonic logging, and
b low-frequency (long
wavelength, kL) wavelet as
that used in the normal
(S 10 cm) (L 60 m)
seismic reflection
surveying

The concept may be extended to cover the


ET a0 2 a1 2 a2 2
Vibroseis source function which is not impulsive
source, but long wavelet (about 2050 s), The energy build-up function furnishes the
reduced to the equivalent impulsive form in the means for dening the delay properties of
processing stage. The seismic wavelet normally wavelets. A wavelet that has its energy concen-
detected by geophones in land surveying, trated at the front of the wavelet is dened to be a
expresses the particle vibration velocity. It is, minimum-delay wavelet, while a wavelet that has
typically, made up of few cycles with duration of its energy concentrated at the end is dened to be
about (4050) ms. In a medium of velocity of a maximum-delay wavelet. According to this
2500 m/s the wavelet dominant wavelength for concept, the seismic source wavelet is a mini-
frequency of 40 Hz, is about 60 m. The corre- mum delay wavelet. More elaborate treatment of
sponding wavelength for 20 kHz (as used in this concept and its applications are found in
well-logging), it will be around 10 cm. Thus, one (Robinson 1967).
can conceive the difference between the wavelets
shown by well-logs and those displayed by nor-
mal seismic traces (Fig. 8.12). 8.4 Sampling and the Digital
Function
8.3.2 Energy and Delay Properties In nature, all geophysical functions are of ana-
of Wavelets logue mode. These functions can be converted
into digital by a process called digitization or
The wavelet energy is expressed by its energy sampling. In seismic surveying this is done by
cumulative function. For a digital wavelet of the recording system as we have explained ear-
sample values (ai), its energy content is expres- lier. The analogue-to-digital conversion (A-D
sed as the sum of the squares of its sample-values conversion) consists of two steps; the quantiza-
(amplitude values). Thus, for a wavelet made up tion (measuring the function values at regular
of three samples, say, (a0, a1, a2,), its total energy intervals) and then conversion into digital num-
(ET) is dened to be the sum of the squares of the bers. This process constitutes what is commonly
wavelet sample values, that is: known, the sampling process.
8.4 Sampling and the Digital Function 179

8.4.1 The Digital Function 8.4.2 The Sampling Process

Functions of geophysical signals can be repre- The process of converting an analogue function
sented either as continuous function (analogue into digital form (A-D conversion) is normally
function) in which the function is dened at all of referred to as (digitization or sampling) process.
the points along the abscissa or dened at discrete It involves two-steps; sample denition followed
points, normally regularly spaced values (digital by quantization and coding in which the sample
function). Considering time-domain functions, value is determined and converted into digital
the analogue function is normally quoted as s(t), number. The output of the sampling process is
and the digital function as s(n  t), where t is the then stored on a suitable digital storage device or
sample spacing along the time-axis, normally entered in a certain digital data processing
called the sampling period. The two forms of operation. In the language of mathematics, the
functions are shown in Fig. 8.13. A-D process is represented as follows:

s(t) by sampler s(n. t )

where, (Dt) is the sampling period and (n), the


sample order number. The sample value is given
The main elements which dene the digital
by the function value, s(n  Dt).
function are the function values (sample values),
The reverse process, digital-to-analogue (D-A)
the sample interval (sampling period), and sam-
conversion, can be carried out by one of the
pling frequency which is reciprocal of the sampling
methods (staircase approximation, linear interpo-
period. For example, seismic traces are normally
lation, or by curve tting), as shown in Fig. 8.14.
recorded at 2-ms sampling period, and processed at
Another more elaborate method based on (sin
4-ms sampling period. The corresponding sam-
x/x) computations can be applied for the con-
pling frequencies are 500 and 250 Hz respectively.
version which is based on summing weighted
and shifted (sin c) functions (Bth 1974, p. 146).

Fig. 8.13 Denition of


the analogue and the digital s(t) s(n.t)
functions

t t
0 0
analogue Function, s(t) digital Function, s(n. t)

Fig. 8.14 Means of


converting a digital
function to analogue mode
(D-A conversion)

staircase approximation linear interpolation curve fitting


180 8 The Seismic Reflection Signal

8.4.3 Representation Methods of the Digital Function

A digital function can be represented in one of the following forms:


(i) Functional Representation
It is represented as a mathematical digital function. Example: 4-sample function:

s(n.t) = +17 for n = 1, 4


-15 for n = 2, 3
0 for n = elsewhere

(ii) Tabular Representation


It is represented as a table of ordered values. Example: 9-sample function:

s(n. t) = 5 3 1 5 3 2 0 7 12

n = 3 2 1 0 1 2 3 4 5

(iii) Time-series Representation


In this case, the sample at zero time is indicated by an arrow. If the arrow is not present it
means that the rst sample is the zero time. Putting dots () at the beginning and at the end of
a series implies that the series is an innite series.

s(n. t) = { 2 5 3 1 3 0 2 5 7 }

s(n. t) = { 2 5 3 1 3 0 2 5 7 }

s(n. t) = { 2 5 3 1 3 0 2 5 7 }

(iv) Z-transform Representation


The z-transform is a mathematical transformation process with which a time series is expressed
as a polynomial in (z). Thus, the z-transform of a digital function represented by the time series
(a0, a1, a2, , aN) is (a0 + a1 z + a2 z2 + a3 z3 +  + aN zN), that is:

a0 , a1 , a2 , a3 , ... , aN z-transform a0+a1z+a2z2+a3z3+... + aNzN


8.4 Sampling and the Digital Function 181

The z-transform of a digital function is a function can be considered as a digital function


polynomial in (z) where the constant sampled at zero sampling-period.
coefcients are the sample values of the It turns out that it is possible to safely sample
digital function. The power of (z) repre- a signal (with no loss of information) if the
sents the order-number (sequence number) digitization process is implemented with the
of the sample. The z-transform polynomial optimum sampling period. A theorem is found
f(z) of a time series (a0, a1, a2, , aN) can which mathematically connects that optimum
be written in a compact form: sampling period with the cut-off frequency of the
X
N signal. This is the Sampling Theorem.
f z an z n
n0
8.5.1 The Sampling Theorem
where (n = 0, 1, 2, , N).
As an example, the z-transform of the This theorem (also called Shannon theorem)
digital wavelet an = {3, 2, 1, 4, 7} is states that, it is possible to completely recover the
given by f(z), where: f(z) = 3 + 2z original analogue function from its digital form if
z2 + 4z3 + 7z4. the sampling period (t) is less than half of the
(v) Graphical Representation smallest period (s) present in that function. This
Each sample is represented by an arrow of implies that, the maximum sampling period
length proportional to the sample value. applied, which incurs no loss in information, is

-3 -2 -1 0 1 2 3 4 5 6 7 8

8.5 The Sampling Theorem that sampling which produces at least two sam-
and Aliasing ples per period of the period-component having
the shortest period in that function. The concept
Digitization of a signal implies reading the signal
is shown in Fig. 8.15.
at isolated points which means that the infor-
In the frequency-domain language the theo-
mation of the signal in between these points is
rem statement can be rephrased as, it is possible
permanently lost. The amount of information lost
to completely recover the original analogue
is naturally greater as the sampling period is
function from its digital form if the sampling
larger. In the sampling process, the shortest the
frequency (fs) is greater than twice of the highest
sampling period, the better is the recovery of the
(maximum) frequency (fm) present in the original
details of the function. In fact, the analogue
182 8 The Seismic Reflection Signal

Fig. 8.15 The


mathematical expression
(t = s/2), the condition
required by the Sampling
Theorem

t t

t = /2

analogue function. The highest frequency is fN  fm


also called the cut-off frequency. The
complete-recovery condition can be mathemati- If the sampling process is too coarse
cally represented by: (too-large sampling periods, too- low Nyquist
frequency), then the output will have certain type
Dt  s=2 of distortion, normally referred to as Alasing.
or (in terms of frequency),

fs  2 fm
8.5.2 The Aliasing Phenomenon

There is another term which is intensively There are three categories of sampling; ne
used in connection with signals sampling, called sampling, critical sampling, and coarse sampling,
Folding frequency or Nyquist Frequency, (fN), depending on the magnitude of the Nyquist fre-
which is dened to be half the sampling fre- quency (fN) relative to the cut-off frequency
quency, that is: (maximum frequency component, fm) present in
the original analogue signal. The categories are:
f N f S =2 1=2Dt (i) Fine Sampling (fN > fm)
In this case high sampling frequency is
With the introduction of the Nyquist Fre-
applied where (fN > fm) complying with
quency (also called folding frequency), the con-
the sampling theorem and complete
dition in the sampling theorem can be re-stated as
recovery of the original analogue signal
(For complete recovery of the original analogue
can be obtained, with no distortions.
function from its digital form, the Nyquist fre-
(ii) Critical Sampling (fN = fm)
quency must be greater than the highest fre-
As in the ne-sampling case, no distor-
quency present in the original signal).
tions occur in this case. It is critical, in the
The important conclusion extracted from the
sense that the applied sampling frequency
Sampling Theorem is that, Nyquist frequency,
sets the marginal limit of the Nyquist fre-
used in sampling of an analogue signal, must be
quency below which distortion (frequency
equal or greater than the highest frequency pre-
aliasing) shall take place.
sent in that signal, that is:
8.5 The Sampling Theorem and Aliasing 183

fm

fA

Fig. 8.16 The aliasing phenomenon. Input frequency (fin) = 100 Hz, Sampling frequency = 120 Hz, Nyquist
Frequency = 60 Hz, Aliasing frequency (fA) = 20 Hz

(iii) Coarse Sampling (fN < fm) frequency (fN f). This means that if the input
With coarse sampling (or under-sampling, to the sampling process is (fN + f), the output
as it is sometimes called), those compo- frequency will be (fN f). The two frequencies
nents which are of greater frequencies than (fN + f) and (fN f) are called aliases of each
Nyquist frequency will appear as lower other (Sheriff 1973, p. 4). Thus, for example, a
frequencies in the sampling output. This 100-Hz signal is sampled at sampling frequency
phenomenon, which occurs only when of 120 Hz (fN = 60 Hz), then the output will be:
coarse sampling is applied, that is when 60 (100 60) = 20 Hz, which is the aliased
(fN < fm), is called (aliasing). An example frequency (Fig. 8.16).
of an aliasing case is shown in Fig. 8.16. The general rule governing any sampler
input-output relationship is that any frequency
component higher than the Nyquist frequency
8.5.3 The Aliasing Frequency (fN), present in the signal prior to sampling, is
Computation outputted as alias frequency. The sampling fre-
quency (fS), or the Nyquist frequency fN (=fS/2),
In any sampling process, the frequency compo- is the deciding factor for the possibility of
nents in the input analogue function which are occurrence of aliasing. Cases for the ne, critical,
higher than Nyquist frequency will appear at the and coarse sampling of an input frequency of
sampler output as false components (aliased (125 Hz) are presented in Table 8.3.
components) of frequencies lower than input In Table 8.3 one can notice that, in the rst
frequency. An input frequency component (fin) two cases (belonging to the case of
higher than Nyquist frequency by (f) will ne-sampling), the Nyquist frequency is higher
appear to the sampling system, as the lower than the input frequency and hence no aliasing is

Table 8.3 Cases for ne, critical, and coarse sampling of an input frequency of (125 Hz)
Input fm (Hz) Sampling freq. (Period) Nyquist freq. (Hz) Output freq. (Hz) Output status
1 125 500 Hz (2 ms) 250 125 no aliasing
2 125 333 Hz (3 ms) 166.7 125 no aliasing
3 125 250 Hz (4 ms) 125 125 no aliasing
4 125 200 Hz (5 ms) 100 75 with aliasing
5 125 166.6 Hz (6 ms) 83.3 41.6 with aliasing
6 125 142.8 Hz (7 ms) 71.4 17.8 with aliasing
7 125 125 Hz (8 ms) 62.5 0.0 with aliasing
184 8 The Seismic Reflection Signal

output fN 2fN 3fN 4fN 5fN 6fN


frequency
fout PA

0
input frequency, fin

Fig. 8.17 Input-output relationship, in a sampling process. The zone up to the Nyquist frequency (fN) is the
principal-aliases zone (PA) in which no aliasing occurs

taking place. In the third case (the applied with no aliasing effect. An input fre-
critical-sampling case), Nyquist frequency is quency, lower than the Nyquist frequency, will
equal to the input frequency and again no alias- be outputted with no aliasing effect.
ing occurs. In cases (4, 5, and 6) Nyquist fre- The periodic linear relation of the sampler
quency is less than the input frequency, hence input-output function, shown in Fig. 8.17, can be
aliasing takes place. In the last case (case 7) the used in deriving a general formula that can be
difference (Df) has reached its maximum value used to calculate the output aliased frequency,
which is equal to the Nyquist value giving an given the input frequency and the sampling (or
output of zero-frequency. Nyquist) frequency. With reference to this gure
To give a quantitative measure for the output it can be shown that the output aliased frequency
frequency, use is made of the periodic linear (fout) is related to the input frequency (fin) by the
relation connecting the input frequency (fin) and following relationship:

fout = |fin - [k + {1-(-1)k}fN|


where,
k is integer = fin/fN

output frequency (fout), as shown in Kanasewich The integer (k) is calculated from floor
(1973, p. 89). Because of the feature that the truncation of (fin/fN), which is symbolically
spectrum-part above the Nyquist frequency can represented as shown above, that is,
be folded back about the Nyquist frequency, k bf in =f N c
aliasing is sometimes called (folding) and the For application, let us nd the output aliased
aliased frequency, called folding frequency. The frequency for the input frequency, (fin = 782
(fin fout) relation is re-produced in Fig. 8.17. Hz), sampled at 2-ms sampling period, i.e. at
This gure is helpful in determining the sampling frequeny (fS = 1/0.002 = 500 Hz). The
inputoutput relation for any sampling process, Nyquist frequency, fN = 500/2 = 250 Hz, and
once the sampling frequency (fS), and hence, the the integer (k) is obtained from truncation of
Nyquist frequency (fN) is known. In the zone (0 (fin/fN = 782/250 = 3), hence, fout = |782
fN), the output frequency (fout) is the same as [3 + {1 (1)3}  250| = 218 Hz. Similarly,
the input frequency. This region of the spectrum, it is possible to calculate the aliased frequency of
called the principal aliases section, marks the any other input sampled signal having frequency
limits of the sampling frequencies which can be exceeding the Nyquist frequency. As examples
8.5 The Sampling Theorem and Aliasing 185

for this method of calculation, let us work out the 8.5.5 Aliasing in the Frequency
aliased frequencies for the input frequencies Domain
(fin = 457, 557, 957, 1957) Hz, each of which is
sampled at 2 ms sampling period. The output The analogue function can be looked upon as a
aliased frequencies for any one of these fre- digital function sampled at zero-sampling period,
quencies will be (fout = 43) Hz. which will have an innitely-long spectral repe-
tition cycle. This is equivalent to saying that its
spectrum is of non-repetitive nature in case of an
8.5.4 Effect of Sampling on Signal analogue signal. When the analogue is converted
Spectrum into digital form the amplitude spectrum is
repeated over the frequency axis at
The direct effect of the sampling process on the repetition-spacing equal to the sampling fre-
frequency spectrum of a signal is generation of quency (fS). The aliasing problem occurs when
repeated spectrum-replicas of the original ana- the cut-off frequency (fm) of the analogue signal
logue signal. Theoretical computations have exceeds that of the Nyquist frequency (fN), that is
proved that Fourier transform of a digital function when (fm > fN). In this case, the time-domain
is itself periodic with repetition interval equal to aliasing becomes spectrum overlapping in the
the sampling frequency used in the sampling frequency domain with the consequence of
process (Kansewich 1973; Kulhanek 1976). Thus, spectrum distortion due to the resulting
after sampling of an analogue signal, its spectrum spectrum-interference. Aliasing, in the frequency
gets repeated over the frequency axis at regular domain, is, for this reason, called spectrum
spacing which is equal to the sampling frequency contamination (Bth 1974, pp. 148151).
(1/t), where (t) is the sampling period. The important point here is that when the
The lower the sampling frequency the smaller input analogue signal contains frequency com-
the spacing becomes. With too-low sampling ponents higher than the Nyquist frequency
frequency (too large t), the spectrum spacing (fm > fN), these high frequencies will fold over
gets smaller and when Nyquist frequency and add to those frequencies existing below the
becomes less than the cut-off frequency of the Nyquist frequency. The outcome of the process
signal (fN < fm), the repeated spectra overlap and in this case is spectrum distortion in the overlap
spectrum distortion occurs. zone as shown in Fig. 8.19.
The spectrum behavior caused as a result of
digitization is shown in Fig. 8.18.

Fig. 8.18 Digitizing effect time-domain frequency-domain


on signal spectrum:
Generation of repeated
spectrum-replicas at (1/t)
spacing along the
frequency axis
fm fm
analogue function single spectrum

1/t
digital function repeated spectrum
186 8 The Seismic Reflection Signal

Fig. 8.19 Spectrum


distortions resulting from (a) (b)
aliasing effect amplitude spectrum of amplitude spectrum
input analogue signal after signal digitization

freq. freq.
fN fN

8.5.6 The Remedy for the Aliasing than the Nyquist value, and hence, aliasing is
Effect avoided.
In summary, in order to avoid aliasing, there
To avoid aliasing and its distortion consequences are two alternative methods:
a procedure should be taken such that the con-
(i) Applying higher sampling frequency (small
dition (fN > fm) is restored. To achieve this state,
sampling period) up to the value with which
one can either delete those frequencies which are
the Nyquist frequency becomes greater than
greater than Nyquist frequency prior to the dig-
the cut-off frequency of the input analogue
itization process, or increase the sampling fre-
signal.
quency in such a way as to get a Nyquist
(ii) Applying a high cut lter (the anti-alias
frequency of greater value than the cut-off fre-
lter) to remove those high frequencies
quency of the original analogue signal.
(from the original analogue signal) which
In practice this is done by application of a
are higher than the Nyquist frequency
suitable high-cut lter (usually referred to as
before starting the sampling process. The
anti-alias lter), to the input seismic data prior
two methods are explained in (Fig. 8.20).
to the sampling process. In effect, the applica-
tion of the anti-alias lter makes the cut-off In seismic reflection data-recording, the sam-
frequency of the input analogue signal, lower pling period is normally set at 2 ms value. This

Fig. 8.20 The two


methods used as remedies
for the aliasing effect
spectrum of the analogue signal

spectrum overlapping (aliasing)


The Two Remedies:
(1) Use of small sampling period (increasing Nyquist frequency)

(2) Removing components higher than Nyquist (anti-alias filtering)


8.5 The Sampling Theorem and Aliasing 187

makes the Nyquist frequency to be 250 Hz. This vertical resolution and horizontal resolution (ex-
means that with this Nyquist value, no aliasing plained here-below).
effect is expected since the maximum frequency
(signal cut-off frequency) is expected to be far
below (250 Hz). However, all recording systems 8.6.1 Vertical Resolution of Seismic
are equipped with the anti-aliasing lter that can Signals
be applied when it is required. The anti-aliasing
lter is normally applied whenever re-sampling Vertical resolution of seismic reflection events, is
of the data is carried out. Very often, in seismic dened as the minimum vertical distance
data processing, the recorded digital data is between two interfaces that give two distict
re-sampled from 2 to 4 ms sampling period for reflection events on a seismic section. It is basi-
economic motives. cally governed by the wavelength of the seismic
signal. The shorter the wavelength (i.e. the higher
the frequency) the greater the vertical resolution.
8.6 Signal Resolution In addition to the frequency factor, depth and
and Resolution Power reflector spacing have signicant effects on the
resolution. The vertical resolution is governed by
Resolution is dened as the ability of distin- the ratio of the depth separation-distance of the
guishing individual objects gathered to gather in reflectors (z) to the wavelength (k) of the
one group, or the details of shape changes of an incident seismic signal. The lowest limit (reso-
irregularly shaped object. In the eld of seismic lution limit) of this ratio; (z/ k) is found to be
exploration, the seismic resolution is the ability (1/4) (Sheriff and Geldart 1995, p. 174). This
of recognizing two adjacent seismic events as means that the reflector separation must be more
distinct two events and not as one blurred event. than quarter of the wavelength (z > k/4). This
Resolution power can be measured by the also means that the image separation (events
minimum separation distance between two seis- separation measured on a seismic stack section)
mic events that can be resolved as two distinct should be more than half a period (t > s/2) in
features on the seismic section. Obviously the order to be distinctly resolved (Fig. 8.21).
sharper the reflection wavelet, and higher Adoption of the (k/4) criterion for the resolu-
signal-to-noise ratio, the better the resolution tion limit, implies that the two events reflected
power will be. The term (resolution power) is from two neighboring interfaces are separated by a
used to imply ability of detecting and bringing to half cycle, which means that depth separation (z)
vision a certain seismic event between two neighboring interfaces greater than
In seismic work, we are concerned with two (k/4) will lead to minimum destructive
types of resolutions of seismic reflection events:

Fig. 8.21 The seismic


vertical resolution.
Minimum reflector
separation (z) is equal to
(k/4)
R1
z = /4
R2

z = v. t/2 = v. (/2)/2 = /4
188 8 The Seismic Reflection Signal

interference between the reflected waves from the earth high-cut lter with depth, both are
two interfaces, causing resolution deterioration. leading to increase of the wavelength and
hence decreases the resolution power.
(i) Factors Affecting Resolution
Application of high-cut lters (as
In general the higher the frequency content
application of the anti-aliasing lter),
of the seismic trace the better is the reso-
which lead to attenuation of the high
lution power. Well logs (wireline logs)
frequencies of the signal and hence
have greater resolution power than seismic
result in lowering the resolution power.
traces since well logs are generated by high
(ii) The problem of Thin Beds
frequency sources. These logs can resolve
A special case, related to the subject of
beds on centimeter-meter scale while seis-
resolution which brought appreciable
mic reflection records cannot resolve
attention by geophysicists, is the problem
so-much detailed variations. Reflection
of resolving thin beds. Two reflectors
survey data can resolve reflectors at
spaced by less than quarter of a wave-
depth-separation of about 10 m at its best.
length, have reflection responses depend-
The main factors affecting resolution are
ing on the layering model. A layer is
reflector spacing, reflector depth, and
regarded as a thin layer when its thickness
reflection signal frequency. Closely spaced
is less than a quarter of the dominant
reflectors cause interferences of reflected
wavelength (Sheriff 2002, p. 353).
waves which lead to loss of resolution. The
Consider a thin bed of thickness of (k/4)
resolution-power is generally decreasing
and of velocity (V2), sandwiched between
with depth for the following reasons:
two layers of velocities (V1 and V3) where
Earth lter which is cutting high fre-
(V1 = V3 < V2), the wave reflected from
quencies, that is cutting short wave-
its top and that from its base will interfere
lengths. Thus for depth of 800 m, say,
constructively producing a high-amplitude
velocity of 1000 m/s, and frequency of
reflection, forming what is normally
100 Hz, the wavelength will be 10 m
referred to as the thin-bed effect or tuning
and the resolution becomes only 2.5 m.
effect. If, on the other hand, the thin-bed of
However, when depth is 3000 m,
velocity (V2) found between two layers of
velocity of 4000 m/s and frequency of
velocities (V1 and V3) where (V1 < V2 <
25 Hz, the wavelength will be 160 m
V3), destructive interference will result
and the resolution becomes 40 m. In
(Sheriff and Geldart 1995, p. 174). These
general resolution gets less (poorer
two models are shown in Fig. 8.22.
resolution) with increasing depth due
Vertical resolution is always improved
to the effect of the earth high-cut lter.
with higher seismic frequencies. But, due
Increase of velocity due to compaction
to the earth ltering effect, frequencies get
and decrease of frequency due to the
lower with increase of reflector depth.

(a) (b)
V1 = V3 < V2 V1 < V2 < V3

V1 V1

V2 V2

V3 V3

Fig. 8.22 Two models of a thin bed, having interval velocity (V2). Model A (V1 = V3 < V2) and model B
(V1 < V2 < V3)
8.6 Signal Resolution and Resolution Power 189

Table 8.4 Estimates of the minimum depth interval resolved, corresponding to thre values of sampling period
Sampling period (ms) Highest frequency (Hz) Wavelength (k) (m) Min. depth-Interval (z = k/4) (m)
1 500 5 1.25
2 250 10 2.50
4 125 20 5.00

Consequently, vertical resolution gets taken as a measure of horizontal resolution power


poorer with increasing depth. on un-migrated seismic data (Sheriff and Geldart
Decrease of frequency (that is increase of 1995, p.177). The radius of the rst Fresnel zone is
wavelength) is leading to decrease of the found to be function of frequency, velocity, and
ratio (z/k) below the limiting value of travel-time of the seismic reflection wave.
(1/4). For example, an incident wave of
(velocity = 1200 m/s, frequency = 40 Hz,
wavelength = 30 m), the minimum depth 8.6.3 Fresnel Zone Formula
interval between two reflectors to be
resolved will be 7.5 m and for a second case The Fresnel Zone concept was originally devel-
of (velocity = 3000 m/s, frequency = oped in connection with the physics of Light.
10 Hz, wavelength = 300 m), the minimum According to this concept, a beam of light,
depth interval between two reflectors to be incident on a reflector will illuminate a limited
resolved will be 75 m. area of its surface, and reflection will take place
(iii) Role of the Sampling Period in Vertical from the area of that surface and not from a point.
Resolution In the same way an incident seismic wave-front
Use of a sampling period (t) which gives a would be reflected from a surface-area of the
Nyquist frequency (1/2  t) higher than the reflector. Thus, the incident seismic energy is
cut-off frequency of the highest frequency reflected from a dened area of the reflector
component of the seismic signal, will avoid surface, rather than from a point (Sherrif 1977).
aliasing effect. Examples of minimum depth Fresnel Zone concept is based on the
interval (z) for a layer having velocity of assumption that all of the reflected energy con-
2500 m/s is presented in Table 8.4. tained in the positive half of the incident
This table shows that the resolution is wave-front, is contributing in illuminating of the
reasonably good even if the sampling reflection area, which is circular in shape in case
period is as large as 4-ms value. This is of vertical incidence. The reflected energy comes
adequate for the shallow layers where the from the area that is affected by the positive half
velocity is normally less than 2500 m/s. of the cycle of the incident wave. The incident
and frequency is less than 125 Hz. energy will be reflected from the part of the
reflector which is within the half cycle following
the reflection onset (Sheriff 1980). The energy
8.6.2 Horizontal Resolution reflected from this zone is constructively inter-
of Seismic Signals fering to make up the reflection event. In terms of
wavelengths, this occurs for all of the energy
Horizontal resolution concerns the ability of rec- reflected within quarter wavelength of the inci-
ognizing two neighboring reflecting points (on a dent wave (Fig. 8.23).
horizontal reflector) as two distinct points and not Lateral resolution is determined by the radius
one point. The minimum separation distance for (R) of the rst Fresnel zone, which is related to
two horizontally-adjacent features, is used as the signals frequency (f), reflection time (T) and
measure for the horizontal resolution. Alongside propagation velocity (v) by an expression derived
with other methods, the rst Fresnel zone is often by (Sheriff and Geldart 1995, pp. 152155):
190 8 The Seismic Reflection Signal

Fig. 8.23 Fresnel zone for


a spherical wave. Two two wave-fronts
wave-fronts separated by /4 apart surface
quarter wavelength (k/4).
R, Fresnel-zone radius
Z
Z
/4
R reflector
first Fresnel zone

understandable since migration has the effect of


R vT=4f1=2 for incident spherical wave
collapsing Fresnel Zones. For this reason, use of
Fresnel Zones in the study of horizontal resolu-
and tion, is applicable only on unmigrated data.

R vT=2f1=2 for incident plane wave


8.7 The Common Numbering
These functions show clearly that the radius of Systems
the rst Fresnel zone (R) is function of the
propagation velocity, reflection time, and fre- To express a sequence of entities, a certain
quency. The variation is direct with velocity counting system is used in which special symbols
(v) and square root of reflection time (T), and have been adopted. These symbols differ with
inverse with square root of frequency. The different languages. The familiar system we are
functions also show that the area of the zone is normally using is the decimal numbering system,
inversely proportional to frequency (f). Thus, in which we use ten different symbols (0, 1, 2, 3,
Fresnel Zone is larger for low frequency com- 4, 5, 6, 7, 8, 9). To continue numbering beyond
ponents than for high frequencies. the number (9), certain combinations of these
It is important to note that Fresnel Zone radius symbols are used. Examples of numbering sys-
cannot be used as a measure for horizontal res- tems are presented in Table 8.5.
olution on migrated seismic sections. This is

Table 8.5 Numbering 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 Radix = 10 (decimal system)


systems. The decimal and
binary are the most 0, 1, 2, 3, 4, 5, 6, 7, 8 Radix = 9
commonly used numbering 0, 1, 2, 3, 4, 5, 6, 7 Radix = 8 (octal system)
systems
0, 1, 2, 3, 4, 5, 6 Radix = 7
0, 1, 2, 3, 4, 5 Radix = 6
0, 1, 2, 3, 4 Radix = 5
0, 1, 2, 3 Radix = 4
0, 1, 2 Radix = 3
0, 1 Radix = 2 (binary system)
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A Radix = 11
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B Radix = 12


0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F Radix = 16 (hexa-decimal system)
8.7 The Common Numbering Systems 191

This table shows that it is always possible to Decimal 105 104 103 102 101 100
devise a new numbering system, but the most system
suitable for us is the decimal system and most Binary system 25 24 23 22 21 20
suitable for computers (computations and stor- Octal system 85 84 83 82 81 80
age) is the binary system. Hexa-decimal 16 5
16 4
16 3
16 2
16 1
160
system

8.7.1 Numbering System Concept

In any numbering system, a number is repre- 8.7.2 Applications of the Concept


sented by a suitable sequence of digits, where
each digit has its own form, position, and value. The most important application of the numbering
Usually, the digits are written from right to left, system concept is done in system-to-system
where the least-value digit is written at the right conversion. The conversion process may be
end and the greatest-value at the left end. These divided into the following functions:
are called the (least-signicant digit) and
Conversion to decimal,
(most-signicant digit) respectively. Number of
Conversion to binary,
symbols used in the decimal system (called,
Conversion to octal, and
Radix) is (10). Other numbering systems based
Conversion to hexadecimal.
on the same rules used in the decimal system, but
with different radix-values, can be devised. The These functions shall be presented with
number, in any numbering system can be repre- examples to clarify the procedure in each case.
sented as a mathematical sequence, represented
as follows:

[ Number ] = + + E R4 + D R3 + CR2 + BR1 +AR0

where R is the radix, and the letters (A, B, C, ) 8.7.2.1 Conversion to Decimal System
are the digits of the number. Conversion is done by applying the mathematical
The following table shows the concept sequence that denes the numbering system
applied to the four most commonly applied concerned. Examples of the conversion process
numbering systems: are given in the following examples:
192 8 The Seismic Reflection Signal

[Number] = + + E R4 + D R3 + C R2 + B R1 +A R0 general case

[5708]D = 5 x 103 + 7 x 102 + 0 x 101 + 8 x 100 = [5708]D decimal number (D)

[7104]O =7 x 83 + 1 x 82 + 0 x 81 + 4 x 80 = [580]D octal number (O)

[1101]B = 1 x 23 + 1 x 22 + 0 x 101 + 1 x 20 = [13]D binary number (B)

[190B]HD = 1 x 163 + 9 x 162 +0 x 161 +11 x 160 = [4353 ]D hexa-decimal (HD)

8.7.2.2 Conversion to Binary and Octal Systems


Conversion from decimal to binary is done by successive divisions by (2) and by (8) for conversion to
octal. Conversion of the decimal fraction to binary or to octal is done by successive multiplications by
(2) and by (8) respectively. The processes are explained in the following examples:

from decimal to binary


integer fraction
47 1 0.593 0
23 1 1.186 .
11 1 0.372 1
5 1 0.744 0
/ 2 = x 2 =
2 0 1.488 0
1 1 0.976 1
0 0

from decimal to octal


integer fraction
47 7 0.593 0
5 5 4.744 .
0 5.952 4
/ 8 = 7.616 x 8 = 5
4.928 7
4
8.7 The Common Numbering Systems 193

In these examples, we found:

47 D 101111 B 57 Oct

and

0:593 D 0:10010 B 0:4574 Oct

8.7.2.3 Conversion from Binary to Octal and to Hexa-Decimal


The octal number is separated into 3-digit groups starting from the point preceding the fraction part of
the number. Each group is substituted by the corresponding octal number. In conversion to
hexa-decimal number, the same procedure is followed except that the group is now made up of 4
digits instead of 3 digits. Again each group is substituted by the corresponding hexa-decimal number.

equivalences for equivalences for


octal conversion Hexa-Decimal conversion

0 (=000) 0 (=0000) 8 (=1000)


1 (=001) 1 (=0001) 9 (=1001)
2 (=010) 2 (=0010) A (=1010)
3 (=011) 3 (=0011) B (=1011)
4 (=100) 4 (=0100) C (=0100)
5 (=101) 5 (=0101) D (=1101)
6 (=110) 6 (=0110) E (=1110)
7 (=111) 7 (=0111) F (=1111)

Examples:
[ 001 001 010 001 . 100 101 ]B = [ 1121.45 ]O

[ 0010 0101 0001 . 1001 0100 ]B = [ 251.94 ]HD

It is to be noted here that there is yet another less-common numbering system called the
binary-coded decimal (BCD) numbering system. In this system, each of the decimal digits is sub-
stituted by the corresponding group of the binary digits. Each of these groups consists of 4 binary
digits, as in the following example:

[ 635 ]D = (6 ) (3) (5)


[ 635 ]D = [0110 0011 1001 ]BCD
194 8 The Seismic Reflection Signal

8.7.2.4 Addition and Subtraction of subtracted (subtrahend) into its compliment by


Binary Numbers changing the zeros to ones and the ones to zeros
The same approach used in addition and sub- and adding the digit (1) to the resulting number
traction of numbers in the decimal system, is (the Twos Compliment number). Subtraction is
applied in case of the other numbering systems. replaced by addition process. This is the proce-
The rule of (Carry) of the multiples of 10 is dure normally followed by the digital computers.
applied. For the Binary numbers, the rules to be
remembered are the following:

for addition for subtraction


0+0=0 0 - 0=0
1+0=1 1 - 0=1
0+1=1 1 - 1=0
1 + 1 = 10 10 - 1 = 1

Examples for Addition:


decimal octal binary
5 5 0101
+7 +7 + 0111
+13 +15 +1101
[ 25 ] D [ 231 ] O [ 11001 ] B

Examples for subtraction:


decimal octal binary
25 31 11001
- 12 - 14 - 1100
[13 ]D [15 ]O [1101 ]B

There is another method to do the subtraction To clarify the process let us take the example of
process, and that is by use of what is called subtraction of the number [+53]D from the number
(Twos Compliment method). The subtraction [+67]D, assuming that the subtraction process is
process starts with converting the number to be done by an 8-bit computer. This is done as follows:
8.7 The Common Numbering Systems 195

[ 67 ]D = [0 1 0 0 0 0 1 1 ]B
[ 53 ]D = [0 0 1 1 0 1 0 1]B

Twos Compliment of [0 0 1 1 0 1 0 1]B


= [1 1 0 0 1 0 1 0]B + 1 = [1 1 0 0 1 0 1 1]B

Subtraction becomes addition of the two numbers, thus:


[ 67 ]D = 0 1 0 0 0 0 1 1
[ 53 ]D = 1 1 0 0 1 0 1 1 +

By summing we get [14 ]D = [1 0 0 0 0 1 1 1 0]B

For an 8-bit computer, the digit (1) appearing the decimal numbering (1, 2, 3, ,). Note that after
after the 8th digit will be outside the computer the digit (9) in the decimal system comes (10) and
binary-word limit and so will be neglected and the after the digit (F) in the hexa-decimal system
result will be the 8-digit word (1110) which is [14]D. comes (10), and after the digit (7) in the octal
system comes (10), and after the digit (1) in the
binary system comes (10). This means that, after
8.7.3 Counting in the Different the complete set of digits of each system comes
Numbering Systems (10), then counting continues. The following table
shows this method applied for the hexa-decimal,
Here is a table of different numbering systems decimal, octal, and binary systems.
(hexa-decimal, decimal, octal, and binary). Each
column contains the numbering corresponding to

Hexadecimal Decimal Octal Binary

0 10 20 30 . 0 10 20 30 . 0 10 20 30 . 0 10 100 110 .
1 11 21 31 . 1 11 21 31 . 1 11 21 31 . 1 11 101 111 .
2 12 22 32 . 2 12 22 32 . 2 12 22 32 . .
3 13 23 33 3 13 23 33 3 13 23 33
4 14 24 34 4 14 24 34 4 14 24 34
5 15 25 35 5 15 25 35 5 15 25 35
6 16 26 36 6 16 26 36 6 16 26 36
7 17 27 37 7 17 27 37 7 17 27 37
8 18 28 38 8 18 28 38
9 19 29 39 9 19 29 39
A 1A 2A 3A
B 1B 2B 3B
C 1C 2C 3C
D 1D 2D 3D
E 1E 2E 3E
F 1F 2F 3F
The Seismic Processing Tools
9

From mid 1950s till mid 1960s a jump in the In order to extract the useful message (geological
seismic data processing took place. Within this information) from these recorded data, these
interval, the analogue magnetic-tape recording seismic traces are subjected to a set of analysis
and processing, followed by the introduction of procedures, using certain processing tools. These
digital technique, were introduced. A third and tools are numerical analysis techniques applied
equally important factor that boosted the seismic on the input seismic traces. The most common
data processing is the adoption of concepts bor- tools applied in seismic data processing are:
rowed from the communication theory which
Fourier Analysis
considered the seismic wavelet as a travelling
Correlation functions computations
signal similar to the electromagnetic signal.
Convolution (frequency ltering)
The input data to the processing system con-
Deconvolution (inverse frequency ltering)
sists mainly of the digital seismic traces, recorded
Frequency- and velocity ltering
by the eld recording system. The fundamental
Equalization (trace scaling)
principle on which processing is based upon, is
Sample Editing (time-shifting, sign-changing,
that the seismic trace is considered to be as a
sample zeroing)
digital signal of value (amplitude) which is
function of the reflection travel time. In order to These processes form the basic processing
extract the useful information the seismic signal tools in the hand of the geophysicist, ready to be
(seismic reflection wavelet), a number of math- used in processing of seismic reflection data.
ematical and statistical processes, are applied.
These processes are considered to be the tools 9.1.1 The Sine Function
employed to do the required analyses.
The building brick of any physically realizable
function is the sine function. According to Fourier
9.1 The Seismic Processing Tools theorem, a time- or distance-function, is made up
of a sum of innite number of sine-functions. Due
The raw seismic data acquired in the eld, which to its important role in processing in general, and
form the input to the processing system, are in Fourier analysis in special, an introductory
normally digitally recorded traces on special treatment of the sine function shall be given
magnetic tapes. The reflection wavelets received here-below, preceding the discussions of the
by a surface-positioned detection-system (the processing tools listed above.
geophone group) are digitally recorded as func- One way to generate a sine function is by using
tion of recording time forming the seismic trace. a geometric approach. Let us consider a unit circle

Springer International Publishing Switzerland 2017 197


H.N. Alsadi, Seismic Hydrocarbon Exploration, Advances in Oil
and Gas Exploration & Production, DOI 10.1007/978-3-319-40436-3_9
198 9 The Seismic Processing Tools

P1 P1 360
P

P2 P0 P0 P2
O R 90 180 270 360 450

P3 P3

Fig. 9.1 Generation of the sine function (sin h)

of radius equal to unity, in which the radius is relation between the two unit systems; degrees
rotating with constant rotation speed about the and radians:
circle centre. At a given time, the radius (OP in
Fig. 9.1) makes an angle (h) with the x-axis, and 360 2p rad
its tip, point (P) has its projection on the x-axis
represented by point (R). By denition, the sine Given (p = 3.1415927), the relation between
function (sin h) has a value given by the length the two systems is that 1 rad = 57.2958, or
(PR). Now, as the radius rotates about the centre 1 = 0.017453 rad. The argument of sine func-
point (O) the length, PR (=sin h) varies with the tions (h, in this example) is normally quoted in
angle (h). By drawing the value of the PR-length radians.
as function of (h), we obtain the curve for the sine
function (sin h), as it is shown in Fig. 9.1.
9.1.3 Parameters of the Sine
Function
9.1.2 The Degrees-Radians
Relationship The general form of the sine function is that it is
function of angle (h). Other common forms of
Referring to Fig. 9.1, the point (P) is rotating at this function are: function of time s(t) and
constant speed and the generated curve is made function of distance s(x), thus:
up of repeated cycles corresponding to the
repeated rotation cycles of the point P around the sh a sinh
circle circumference. The generated (sin h curve)
repeats regularly every 360. The complete angle or,
swept by the rotating arm is from (h = 0) when P
st a sin2pf t t tp ; f t 1=s
is at location (P0) to (h = 360) when P is back at
location (P0) passing through all of the points or,
(P0, P1, P2 and P3).
It should be noted here that angles can be sx a sin2pf x x xp ; f x 1=k
expressed in radians rather than in degree-units.
An angle (h) in radians is dened as the ratio where, (a) is amplitude, () is phase angle, (ft) is
between the subtending arc divided by the radius temporal frequency, (fx) is spatial frequency, (s)
of the circle in which the angle is central. When is period, (k) is wavelength, (tp) is phase shift in
the arc length is equal to the radius, the angle time units, and (tx) is phase shift in distance
value is 1 rad. Since circumference of a circle is units.
equal to 2p times its radius, the value of the total The maximum value of the sine function is
central angle is 2p. This leads to the important equal to the length of the rotating arm applied for
9.1 The Seismic Processing Tools 199

generating the sine function considered above. arm sweeps a complete central angle (360),
A sine function, represented by a sine curve which is (2p) when expressed in radian units.
(sometimes called sinusoid) is completely When frequency is expressed in terms of rate of
dened by three parameters: amplitude (a), fre- change of the angle (dh/dt), measured in radians
quency (ft or fx) and phase (, tp, or xp). The per second, it is usually called, radian frequency.
amplitude is of positive value (+a) represents a Thus, for constant rate, we can write:
peak and negative value (a) represents a trough
of the given sinusoid. x h=t 2p=s 2pf
The most common form of the sine function is
the form a sin(2pft + tp) quoted as function of where, (x) is the radian frequency (rad/s) and
time (t). The frequency, (f) is expressed in cycles (f) is the cyclic frequency (cycle/s).
per second. The sine function is of innite length The frequency (f) is normally referred to as
extending along the time axis in both of the cyclic frequency, measured in cycles per second, to
negative and positive directions. Part of a sine differentiate it from the radian or angular frequency
function and its three characteristic parameters (x) which is measured in radiance per second.
are shown in Fig. 9.2. Frequency Units
The unit normally used in measuring frequency
is the Hertz (Hz), where 1 Hz is 1 cycle/s.
9.1.4 The Frequency Concept Another closely related measurement-unit
applied in measurement of frequencies is the
Any periodic motion, as the rotation-arm model Octave. Instead of quoting the frequency in its
shown in Fig. 9.1, is specied by two periodicity absolute units (Hertz), it is quoted in frequency
parameters: the frequency and the period of the ratios measured in Octave-units. It is dened as
motion. Frequency is dened to be the number of the number of times (n) with which a frequency
revolutions (cycles) made in one second. The (f2) value gets doubled over a reference fre-
length of the time interval (normally measured in quency (f1). A frequency ratio (f2/f1) is one
seconds or milliseconds) for one cycle, is called Octave if (f2/f1 = 2). In general, the number of
the period. The relation between frequency octaves (n) is related to the ratio (fn/f1) by:
(f) and period (s) is:
f n =f 1 2n
f 1=s n 1=log 2  logf n =f 1
n is number of Octaves
For the rotation arm, shown in Fig. 9.1, the
period (s) is the time interval during which the

Frequency Value in Frequency examples


s(t) ratio octave-units
21 : 1 1 (1020) Hz = 1
tp
octave
+a 22 : 1 2 (1040) Hz = 2
t octaves
-a 23 : 1 3 (1080) Hz = 3
octaves
21/2 : 1 1/2 (1014.1) Hz = 1/2
octave
Fig. 9.2 Part of a sine function, a sin(2pft + tp), and its
characteristic parameters amplitude (a), period (s = 1/f), 21/3 : 1 1/3 (1012.8) Hz = 1/3
and phase shift (tp) octave
200 9 The Seismic Processing Tools

Examples: the time interval between time zero and the


It is common practice that the slope of frequency nearest peak, the value of the phase must be
spectrum is measured in decibel-per-octave units. conned within the range (s/2) and (+s/2), that
Thus, for example, an amplitude spectrum is between (180) and (+180).
dened by four corner frequencies (f1, f2, f3, f4), When the peak falls exactly on time zero, the
the slopes of the spectrum between (f1 and f2) phase is zero. If, however, the peak falls later in
and between (f3 and f4) are normally quoted in time than time zero, the phase is called a lag
decibel per octave, where f2/f1 = f4/f3 = 2. phase and the phase angle, in this case, is given a
A frequency spectrum of this form is given in negative sign. If, on the other hand, the peak falls
Fig. 9.3. before time zero, the phase is described as lead
The frequency bandwidth of such a lter-form phase and the phase angle is given a positive
is considered to be the frequency range between sign. When a trough coincides with the time zero,
the two points located at 6 db level on the lter we get a case where the peaks become at equal
amplitude spectrum as shown in Fig. 9.3. distances from the time zero, and the phase can
be called either (180) or (+180). These de-
nitions are explained in Fig. 9.4.
9.1.5 The Phase Concept Phase expresses the relative position of a sine
wave with respect to the time of origin that is
The phase expresses the position of the cycle of a position relative to time zero. The value of phase
sinusoid in relation to the time zero. The con- may be expressed in fractions of the period or in
vention followed by geophysicists is that the time units (usually in seconds, or milliseconds),
phase shift of a sinusoid is considered as the considering that one complete period (s) is
interval between the time zero and the nearest equivalent to (2p) or (360). Common examples
peak of that sinusoid. As stated above, the of phase measurements are shown in Fig. 9.5.
argument of the sine function is either angle (h),
a sin(h + ), or time (t), a sin(xt + tp). The
terms () and (tp) are representing the phase 9.1.6 Temporal and Spatial
parameter. Frequencies
Phase is measured in degrees when the argu-
ment of the sine function is angle or in time units Frequency (ft) is the number of cycles per second
when the argument is time. It is always possible when the argument of the sine function is time,
to convert phase values from degrees to time s(t). In the x-domain, where the argument of the
units and vice versa using the fact that one sine function is distance (x), the frequency (fx) is
period-length is equal to (360), or to (2p radi- expressed in cycles per meter. The frequencies
ans). Thus if the phase angle is (90) then the (ft) and (fx) are termed temporal and spatial fre-
time-equivalent is quarter of the period and so quencies respectively. The period (s), is equal to
on. Since, by denition, phase is represented by the time interval separating two neighboring

slope slope
(in db/octave) (in db/octave)
filter frequency bandwidth

frequency
f1 f2 f3 f4
Fig. 9.3 Slopes of a four-corner spectrum quoted in decibel (db) per octave units, where f2/f1 = f4/f3 = 2
9.1 The Seismic Processing Tools 201

s(t) s(t)
tp tp

t t
0 0
phase lag = tp phase lead, tp

s(t)
s(t)
tp tp

t t
0 0

zero phase, tp= 0o phase, tp = +T/2 or T/2

Fig. 9.4 Phase terminology

Fig. 9.5 Common sine waves with different


examples of phase-shift, phase degrees period
phase shifts
with their corresponding
terminologies

Phase - 90 + /4
lag

Zero 0 0
phase

Phase + - /4
lead
90

+
180 + /2
Out-of -
phase 180 - /2
202 9 The Seismic Processing Tools

s(t)
s(x)

a a
t x

Temporal sine function s(t) Spatial sine function s(x)

Fig. 9.6 Denition of the period (s) and wavelength (k) of a sine function in the two domains, s(t) and s(x)

peaks, in a time-domain sine function. Likewise, simplest form, this type of function has the form
the distance separating two peaks in x-domain (Richardson 1953, p. 38):
sine function represents the wavelength (k) as
shown in Fig. 9.6. sx; t a sin 2pt=s  x=k
In radians units, we have the temporal relation
(x = 2pft = 2p/s) and the corresponding spatial Using the relationships (x = 2pft = 2p/s) and
relation (k = 2pfx = 2p/k). Another important (k = 2pfx = 2p/k), the function can be written
relation is the one connecting the temporal as:
parameters with the spatial parameters, which is:
sx; t a sinxt  k x
v k=s
where k is constant known as the wave number,
It should be noted that when frequency is v is propagation velocity, and x is the radian
quoted as (f) without the subscript (t), it is nor- frequency measured in radians per second.
mally meant to be the temporal frequency (ft). It is important to note that the wave propa-
gation velocity (v) is connected with both of the
temporal frequency (ft) and wavelength (k) by
9.1.7 Propagating Sine Wave the simple relationship:

v f tk
The rotating arm model (shown above), helped in
generating a one dimensional sine function that
or,
depends on one variable which is angle (h) or
time (t). The time-dependent sine function v x=k
describes the case of stationary (non-propagating)
periodic motion. However, we may have a Using a simple pictorial method, the velocity
two-dimensional sine function s(x, t) describing formula (v = ft k) can be derived. Referring to
the case of a propagating sinusoidal wave. In its Fig. 9.7, the number of cycles (ft) of a
9.1 The Seismic Processing Tools 203

Fig. 9.7 Derivation of the sinx


relation connecting
propagation velocity
(v) with frequency (ft) and
wavelength (k) x

distance covered in one second

v = ft

propagating sine wave measured in 1 s is mul- (iv) The integral of f(t)dt over a complete
tiplied by the cycle length (wavelength, k) gives period is convergent.
the total distance covered in that 1 s, which is the
It so happened that almost all geophysical
propagation velocity (v). Thus (v = ft k).
phenomena, including the seismic signal, obey
Dirichlet conditions and hence can be analyzed
by Fourier Theorem.
9.2 Fourier Analysis and Concept
of Spectra
9.2.1 Fourier Series
In 1807, Joseph Fourier (17681830) presented
his theorem (now identied by his name) which According to Fourier Theorem, any periodic
states that any function such as f(t), satisfying function, f(t) of period (s), satisfying Dirichlet
certain restrictions can be expressed as a sum of conditions, can be represented by the sum of
an innite number of sine waves. In 1809 sinusoidal functions of frequencies which are
Dirichlet formulated the mathematical restrictions multiples of the fundamental frequency of that
(normally referred to as Dirichlet conditions) function. Mathematically, the periodic function
under which the theorem is mathematically valid. f(t) is expressed by the following innite series
The restrictions are: (called Fourier series):

where (x), is the fundamental frequency (angular


(i) The function is periodic, f(t) = f(t + s),
frequency, in radians). That is:
where (s) is the period.
(ii) The function is sectionally continuous, x 2pf 2p=s
with nite number of discontinuities. The constants (a0, an, and bn) which are
(iii) The function is possessing nite number called (Fourier coefcients) can be determined by
of maxima and minima. the nite integrals:
204 9 The Seismic Processing Tools

a0 =

an =

bn =

These three expressions are derived by mul- of frequencies (xn) and amplitudes (an and bn).
tiplying both sides of the Fourier series in turn by The term (a0) represents the DC level of the
(1), (cos nxt), and (sin xt), and integrating with function f(t).
respect to (t) over the period length (s). The
Odd and Even Functions
derivation is based on use of the orthogonality
For an odd function fo(t) in which fo(t) = fo(t)
properties of the sine and cosine functions, which
are (for m and n being integers):

= = /2 for m=n, and (=0, for m n)


and,
= 0, for all m and n

The frequencies (xn) that are multiples of the the coefcients (an) become all equal to zero,
fundamental frequency (x), are called harmonic since in this case:

an = = +

=- +
=0

frequencies or just harmonics. Thus a periodic Similarly, it can be shown that, for an even
function f(t) is made up of the sum of an innite function fe(t) in which fe(t) = fe(t), the coef-
number of harmonics (sine and cosine functions) cients (bn) become equal to zero for all (n).
9.2 Fourier Analysis and Concept of Spectra 205

Thus, when the function f(t) is odd, the approximation for the continuous parts of the
Fourier series will consist of sine terms only and function. This kind of distortions (synthesized-
when it is an even function, the series will consist function overshooting- and oscillation-behavior)
of cosine terms only. is normally referred to as Gibbs Phenomenon
(Fig. 9.8).
Application of the Fourier series analysis of
9.2.2 Gibbs Phenomenon periodic functions is found in many standard
texts on the subject, see for example (Alsadi
As we have stated above, a periodic signal (pe- 1980, p. 111).
riod, s), satisfying other Dirichlet conditions, can
be expressed as a sum of sinusoidal components
of frequencies; n/s, (n = 0, 1, 2, 3, ). The 9.2.3 Fourier Transform
zero-frequency component (n = 0) represents the
DC term. This implies that if we have a signal, As it is presented above, Fourier series is
f(t) dened over the time interval (t = 0 to restricted to periodic functions, but theory can be
t = T), then, according to Fourier Theorem, the extended to cover non-periodic functions. Func-
function within this interval can be expressed as tions of nite length are called transient func-
the sum of the frequency components (n/T) tions. With special integrals, it is possible to
where (n = 0, 1, 2, 3, , ). transform a transient function f(t) to another form
Fourier Theorem states that the signal f(t) will in which it becomes function of frequency F(x),
be recovered exactly from the Fourier series only with no loss of information. The process of
when an innite number of terms are included in converting f(t) into F(x) is called (Fourier
the summation, that is when the integer (n) runs transform) and the reverse process, F(x) into f(t),
from (n = 1) to (n = ). In practice, only a is called (inverse Fourier transform).
nite number (n = N, say) of terms is used in the The transient function can be considered as a
summation. As a result of this truncation of the periodic function of innitely long period. The
series (i.e. when n is nite number), distortion process of transforming a transient function from
shall occur in the recovered (synthesized) signal. its time domain f(t) to the frequency domain
In this case, the sum will shoot beyond the value function F(x) and vice versa, are done through a
of f(t) in the neighborhood of the discontinuities set of integral equations called (Fourier Integral
found in f(t). The overshoot oscillates about the equations) which are more commonly known as
function value with decreasing amplitude as we Fourier Transform equations. The two equations
move away from the discontinuity. Increasing (Transform and Inverse Fourier transform equa-
the number of terms leads to a better tions) are Bth (1974, pp. 3337):

f(t)
f(t)

t t
(few-terms) approximation (more-terms)

Fig. 9.8 Gibbs Phenomenon computed for a rectangular pulse for few- and more-terms truncation of Fourier series
206 9 The Seismic Processing Tools

(Fourier Transform)

(Fourier Inverse Transform)

In general, the function F(x) is complex, similar to Light spectrum in which the Light
consisting of a real part a(x) and imaginary part frequency-components appear when analyzed as
b(x). Thus, F(x) can be expressed as: it passes through a transparent prism. In the
The two functions, a(x) and b(x), known mathematical sense, F(x) is complex function
respectively as cosine and sine transforms, are consisting of the modulus; jFxj called (am-
dened as: plitude spectrum) and the argument (x), called

F() = a() - ib(),

or.
F() = A()ei()
where,

A() = [a2() + b2()]1/2

() = tan -1 [- b() / a()]

Z1 (phase spectrum). The square of the amplitude


ax f t cos xt dt spectrum is the Fourier power spectrum.
1
Fourier spectrum can be presented in one of
the two forms: F(x) = a(x) ib(x) and F
and, (x) = jFxj ei (x), where,

Z1 p
jFxj ax2 bx2
bx f t sin xt dt
x tan1 bx=ax
1

By using Fourier integral transform, it is pos-


sible to transform a function (such as a seismic
9.3 Concept of the Frequency signal) from its original time-domain function f(t)
Spectrum to the frequency-domain function F(x), with no
loss of information. In this process the original
The function F(x) expresses the variation of time domain signal is analyzed into the frequency
amplitude of the frequency components as components (sine functions) constituting that
function of frequency (x). This is, in a way, signal. The produced frequency-domain plot
9.3 Concept of the Frequency Spectrum 207

represents the Fourier spectrum of the analyzed phase-spectra. Through another form of integra-
function. tion (Inverse Fourier transform), it is possible to
recover the original time-domain function from
9.3.1 The Line Spectrum the transformed frequency-domain function. For
complete recovery, this process requires both of
Based on the nature of the input signal (periodic or the amplitude spectrum and phase spectrum. The
non periodic), the computed spectrum is obtained concept of frequency spectra (amplitude- and
as discrete points or as continuous curve. Fre- phase-spectra) computed by Fourier transform
quency analysis of a periodic function involves equations are illustrated pictorially in Fig. 9.10.
determination of the Fourier series coefcients of In summary, it can be stated that
the input function. In this case the output Fourier series analysis of periodic functions
frequency-domain function is plotted as discrete yields line spectra and Fourier transform analysis
lines along the frequency axis forming what is of non-periodic functions yields continuous
commonly referred to as the (line spectrum). spectra.
The line spectrum S1(x) of a sine function
s1(t) = a sinx1t, is represented by one line of
height proportional to the amplitude (a) of the 9.3.3 The Fourier Power Spectrum
sine function, located at frequency (x1). For an
another sine function, s2(t) = a sinx2t of fre- The total energy (E) of a real function f(t), such
quency (x2), will, likewise, be of one line placed as a seismic signal, is generally taken to be
at frequency (x2). Further, the sum of the two proportional to the integral of the square of its
sine functions s1(t) and s2(t) will have a line amplitude. That is:
spectrum made up of two lines placed at (x1) and
Z1
(x2). The principle is shown in Fig. 9.9.
E ff tg2 dt
1
9.3.2 The Continuous Spectrum
It can be shown (Bth 1974, p. 82) that this
Analysis of a non-periodic signal, by the Fourier expression is related to the power spectrum as
transform, gives continuous amplitude- and follows:

Fig. 9.9 Line spectra of


two sine functions, s1(t)
and s2(t) and of their sum
s1(t)
s1(t) + s2(t) S1( )
t
1

s2(t)
S2( )
t
2

s1(t) S1( )
+ +
s2(t) S2( )
t
1 2
208 9 The Seismic Processing Tools

Fig. 9.10 Pictorial time domain function (signal)


representation of spectrum a(t)
analysis. The signal (at top
of gure) is a0
Fourier-transformed into its time (t)
frequency components
shown as amplitude and
phase spectra
freq. components, an(t) amplitude spectrum, A()

a1

a2 A()

a3 a0 a1 a2 a3 a4 a5

1 2 3 5
a4
() phase spectrum, ()
a5

Z1 Z1 (frequency-domain) function. In fact, any real-


2 1 2 izable function f(t) has its own frequency-domain
ff tg dt jFxj dx
2p counterpart, F(x). The two functions f(t) and
1 1
F(x) are normally called (Fourier pair), when f(t)
is Fourier-transformable into F(x), that is,

Fourier Pair
f(t) F()

The main advantage of the two-domain con-


The real quantity jFxj2 is normally con-
cept is that computation is much simplied when
sidered to be the (power spectrum) or energy
working in the frequency domain. The concept is
spectrum of the function f(t). It is important to
very much like logarithms and anti-logarithms
note here, that the power spectrum is square of
domains. Thus multiplication of numbers
the amplitude spectrum.
becomes simple addition in the logarithm
domain. The outstanding example in time series
9.3.4 The Two-Domain Concept analysis is convolution and deconvolution com-
putations. Convolution and deconvolution pro-
With the application of the Fourier transform, it cesses in the time domain become simple
is possible to transform a function from a func- multiplication and division of the corresponding
tion of time f(t), to a function of frequency F(x) functions in the frequency domain.
with no loss of information. This implies that a To take advantage of this property, some
given function can be expressed either as func- processes in data processing are done by, rst
tion of time f(t), called (time-domain function) or Fourier-transforming the signal (such as a seis-
as function of frequency F(x), which is called mic trace) from time domain to the frequency
9.3 Concept of the Frequency Spectrum 209

time domain, Fourier transform frequency domain,


input function, f(t) f(t) F() F()

time domain, Fourier inv. transf. frequency domain,


output function, f(t) f(t) F() computations

Fig. 9.11 Block diagram showing the principle of domain-to-domain transformation

domain, doing the computation in this domain Za Za


ixt
(which is much faster than in the time-domain) Px pte dt eixt dt
and the result is inverse-transformed to the
a a
original time-domain. This is explained as fol- 2 sin ax=x
lows (Fig. 9.11).
Hence,
9.3.5 Spectrum of the Rectangular
Pulse Px 2 sin ax=x 2a sincax=p

The resulting spectrum function, P(x) of the


As an example of application of the Fourier
transient function, p(t), which is a rectangular
integral, let us determine the frequency spectrum
pulse, is found to be of the form (sin px)/px
of a rectangular pulse, p(t) of width (2a), and
where x = ax/p. The value of this function is
height (1). The function is dened as Fig. 9.12:
zero whenever (sin x = 0). This occurs at
pt 1; a\t\a (ax = np). At the origin, however, where x = 0,
the function value is equal to (2a). With the
0; elsewhere
increase of (x), the spectrum function Px
oscillates about the (x-axis) with decreasing
By direct application of the Fourier Trans-
amplitude. The Fourier pair, pt $ Px, is
R1
form, Fx f teixt dt, we get: sketched in Fig. 9.12.
1

rectangular pulse, p(t) amplitude spectrum P( )

p(t) 2a

2a
t
-a 0 +a -2 /a - /a 0 /a 2 /a

Fig. 9.12 The window pair of the rectangular-pulse, box-car time function, pa(t) and its spectrum, sinc-type function,
P(x) = (2sin ax)/x
210 9 The Seismic Processing Tools

In the geophysical literature, this function is pulse rectangle gets narrower, the corresponding
known as the (sinc) function. The function (sinc amplitude spectrum gets wider, and in the lim-
x) is dened as (sinx)/x or (sinpx)/px (Sheriff iting state, the pulse becomes spike-function
2002, p. 320). Fourier transform P(x) of the (impulse function) with innitely wide spectrum
rectangular function (normally called box-car (Fig. 9.13).
function, p(t) is the sinc function, 2a sinc (ax/p). Another important conclusion can be drawn
from this behavior, and that is the spike time
function contains innite number of equal-
9.3.6 Spectrum of the Spike Pulse amplitude frequency components. For this rea-
son the spectrum of a spike signal is often
Mathematical analysis proved that there is an described as being white spectrum.
inverse relationship between the width of the
time-domain function (as a seismic pulse) and its
frequency domain amplitude spectrum. Thus, as 9.3.7 The Dirac-Delta Function
the pulse gets narrower, the corresponding
frequency-domain spectrum becomes wider, and The zero-width pulse (impulse) is called (Dirac
vice versa. This behavior can be readily seen in delta function) and given the symbol d(t). This
case of the rectangular pulse-spectrum Fourier special function is dened as being a compressed
pair. The main lobe of its spectrum extends from rectangular pulse dened to be of width,
(x = p/a) to (x = +p/a). This clearly indicates approaching zero and having a unit area (Bth
that the spectrum has inverse relation to the 1974, p. 52). It is considered as being even
original pulse width (2a). In fact, this is one of function and hence it has a real spectrum func-
the important properties that hold for all Fourier tion which is constant at the value of unity for all
pairs of pulse-shaped functions. frequencies. This means that the Delta function
According to this principle (called the can be synthesized by superposing an innite
reciprocity property) it is readily seen that as the number of sine functions which are in phase at

Fig. 9.13 The reciprocal


time domain, f(t) Freq. domain, F( )
relation between the width
of the time-domain pulse f(t) F()
and that of its amplitude
spectrum

t
0 0

f(t) F()

t
0 0
f(t)
F()
)


0 t 0
9.3 Concept of the Frequency Spectrum 211

Fig. 9.14 Synthesis of an


impulse function from
summing sine functions of
equal amplitude and in +
phase at the origin time,
t=0 +

+
+

t=0

one point (at t = 0) where the components will The signal length (T, say) denes the lowest, or
add up constructively, and destructively else- fundamental frequency (fL = 1/T) in the spectrum.
where as shown in Fig. 9.14. Also, it denes the frequency increment, f (= 1/T)
Dirac delta function, d(t) is not a proper of the spectrum computation. The digitization
mathematical function. It is usually considered as interval, or sampling period (t), on the other hand,
a mathematical concept which possesses its own sets the upper frequency limit of the spectrum. In
mathematical properties. It is dened as: fact, the upper limit is the Nyquist frequency (fN)
which is equal to (1/2t). With consideration of
Z1 these limits, spectrum of an observed signal, of
dt 0 for t 6 0 and dtdt 1 nite length (T) and digitized at (t) interval, is
1 usually computed for the frequencies:

The delta function is even, d(t) = d(t), and it 1=T; 2=T; 3=T; . . .; 1=2Dt
forms the Fourier pair (d(t) $ 1). Further, we have
another important property which states that its Thus, a digital signal made up of (N + 1)
convolution with a function leaves the convolved samples, will be of length (T) given by
function unchanged, that is: d(t) * f(t) = f(t). (T = Nt). By substituting (T/N) for (t), com-
putation frequencies will be:

9.3.8 Frequency Limits of Fourier 1=T; 2=T; 3=T; . . . ; N=2T


Spectra
That is at frequencies, n fL, where n = (1, 2,
Spectra of observed signals are characterized by 3, , N/2).
two main features. The signal prepared for the From this result, we can draw an important
analysis, is of nite length and it is in digital conclusion which states that the number of sig-
form. Both of these features impose limits on the nicant frequency-values taken in computing a
frequency limits over which the spectrum is frequency spectrum of a digital signal, is half the
computed. number of the samples of that signal.
212 9 The Seismic Processing Tools

Finally, it should be emphasized that the length These are expressed in the frequency domain by
(T), controls the (resolution power) of the com- the amplitude- and phase-spectra as shown in
puted spectrum, since the frequency interval f Fig. 9.15.
(= 1/T) is inversely proportional to the length (T). In Fig. 9.15, signal-1 has negative time-phase
(positive phase-angle), while signal-2 is of
zero-phase. The rest of frequency components
9.4 The Phase Spectrum are of negative phase angles. These amplitude
and phase characteristics are shown in the cor-
From the previous paragraph, we learned that the responding frequency domain spectra.
frequency spectrum F(x), of any real time
function is generally a complex function, con-
sisting of the modulus; jFxj called (amplitude 9.4.1 The Zero-Phase Spectrum
spectrum) and the argument (x), called (phase
spectrum). The importance of the phase charac- If all the frequency components are of zero phase
teristics of a seismic signal is that it has direct (symmetrical about time zero), their sum will
influence on the signal shape. Signal amplitude, give a zero-phase wavelet, symmetrical about
on the other hand has a direct effect on the signal zero time. The phase spectrum, in this case, is a
energy. The phase of a frequency component is straight line drawn along the frequency axis at
measured by the time shift of a peak of a cycle zero phase-value. Amplitude and the phase
with respect to time zero, considered to be the spectrum for this case are shown in Fig. 9.16.
start time. Thus, the start time is the time relative Note that the amplitude spectra in the two
to which all phases of the frequency components Figs. 9.15 and 9.16 are the same, but their
are referred to. In general, the phase value may respective time domain signal-shapes are differ-
be negative, zero or positive. ent. This difference in shape is due only to their
Seismic signals, like seismic reflection wave- different phase spectra. Another useful note is
lets, are all of the type of signals which are that the recorded seismic reflection wavelet is
existing only in the positive side of the origin one-sided signal. It cannot be zero-phase since,
time. Such functions (called one-sided functions) by denition, it has no energy before time zero as
consist of the sum of frequency components of one-sided signal demands. However, conversion
different amplitudes and different phase shifts. into zero-phase wavelet can be done by

amplitude spectrum

1 t

2
1 2 3 4

3
+

sum phase spectrum


t=0
Fig. 9.15 Set of frequency components (sine functions) of a one-sided signal, expressed in the frequency domain by
amplitude- and phase spectra
9.4 The Phase Spectrum 213

amplitude spectrum
1 t

2
1 2 3 4

3
+
4
-
zero-phase spectrum
sum=
t=0

Fig. 9.16 Set of frequency components (sine functions) of a zero-phase signal, with its amplitude and phase spectra

application of special computer programs. This frequency. In this case, each frequency compo-
can be done in processing stage to convert the nent will be shifted by the same time shift
one-sided wavelet to a zero-phase wavelet of (Sheriff and Geldart 1995, p. 533). The sum, in
symmetrical shape. This is normally asked for by this case, will be like the zero phase spectrum
interpreters for more clear denition of reflection (symmetrical wavelet), but the symmetry will be
events and easier to follow in the process of about a time later than zero-time. The phase
interpretation especially for structural purposes. spectrum will be a straight line inclined to the
frequency axis (Fig. 9.17).

9.4.2 The Linear-Phase Spectrum 9.4.3 The Constant-Phase Spectrum

A closely associated case is the linear phase The constant phase spectrum is a special case of
spectrum, in which the phase is linear function of linear phase spectrum. The shift is not function of

amplitude spectrum

1 t

2 1 2 3 4
+
3

-
sum = linear-phase spectrum
t=0
Fig. 9.17 Set of frequency components (sine functions) of a linear-phase signal, with amplitude and phase spectra
214 9 The Seismic Processing Tools

frequency but having a constant value. Thus spectrum in this case is straight line parallel to
when the phase value is constant at zero-value, the frequency axis and located at (p) below (or
we get a zero-phase spectrum as shown in above) the frequency axis. The (p) or (+p)
Fig. 9.16. If the constant shift is a non-zero phase of a component implies that it has a trough
value, the phase spectrum is also a straight line, at time zero. The wavelet resulting from sum-
but it is parallel to the frequency axis. For ming such components will be symmetrical and
example if the constant phase shift is at later-than will have a trough at time zero as shown in
the time zero by quarter a period (s/4, say), Fig. 9.19.
which is equivalent to (p/2), the phase spectrum
will be a straight line located at (p/2) below the
frequency axis. In this case, the frequency com- 9.5 Spectra of Observational Data
ponents will all have the constant phase of (p/2)
and their sum will give an inverted-symmetry As far as Fourier analysis is concerned, an
wavelet that has its main peak immediately fol- observational function differs from an analytic
lowing the time zero (Fig. 9.18). function in two main aspects. The observed
If the phases of the frequency components are function is not innite and not continuous as an
all equal to (+p/2) instead of (p/2), the phase analytic function would be. Thus, a signal,
spectrum will be a straight line at (p/2) above the (normally, time function), inputted to the trans-
frequency axis. In this case, all the components formation process, is nite in length and digital
will have troughs immediately following the time in form. Observational functions, which are input
zero, and the wavelet resulting from summing the for spectral analysis, are usually not continuous
components will, likewise, have a trough fol- and not innite in length as Fourier Transform
lowing time zero. integral demands. For this reason, observational
The last interesting constant-phase case is a spectra suffer from the two types of distortion
wavelet having all its component frequencies of which are resulting from signal-truncation and
equal phases, at the constant phase of (p), signal-digitization. These effects shall be dis-
which is the same as (+p). The constant-phase cussed as follows:

amplitude spectrum

1 t

2 1 2 3 4

3
+
4
-

constant-phase spectrum
sum =
t=0

Fig. 9.18 Set of frequency components (sine functions) of a constant-phase (at phase = p/2), with amplitude and
phase spectra
9.5 Spectra of Observational Data 215

amplitude spectrum

1 t

2
1 2 3 4
3

+
4
-
sum =
constant-phase spectrum
t=0
Fig. 9.19 Set of frequency components (sine functions) of a constant-phase (at phase = p, or at phase = +p), with
amplitude and phase spectra

9.5.1 Truncation Effect innity to plus innity. Being computed for


and Windowing signals of nite lengths, spectra of truncated
signals are always suffering from distortion
Normally, a signal of nite length is effectively which gets more severe the shorter the truncation
equivalent to multiplying of innitely long signal length.
f(t) by a rectangular function of nite-length (T, One of the important properties of Fourier
say) and of constant value (usually equal to Transform (the convolution Theorem) states that,
unity). This function which is sometimes called multiplication of two functions in the time
box-car function, or box-car time window, w(t), domain becomes convolution in the frequency
is dened by: domain, and convolution in the time domain
becomes multiplication in the frequency domain.
wt 1  T=2\t\T=2 That is:
0 elsewhere

f1(t) . f2(t) F1() * F2()

Truncating of a signal is effectively done by


The truncated function, ftr(t) is given by multiplying the innitely long signal by a
ftr(t) = f(t)  w(t). This process will isolate box-car function of value (1) over the needed
(truncate) the length (T) desired for the analysis. part of the signal, that is (f(t)  w(t)). By applying
Truncation, which is an unavoidable process, the convolution theorem, the Fourier spectrum
violates Dirichlet conditions which demands that Ftr(x) of the truncated function ftr(t) is therefore,
Fourier Integral must be computed from minus given by F(x) * W(x), that is:
216 9 The Seismic Processing Tools

Ftr x Fx  Wx can be seen (top case in Fig. 9.20) that an


un-truncated signal is equivalent to a truncated
where F(x) and W(x) are the Fourier transforms signal using an innitely long truncation rectan-
of f(t) and w(t) respectively. gular function (window). Such a window has an
Since W(x) = (2/x) sin (xT/2), we can impulse-form spectrum. Theory shows that con-
write: volution of an impulse with a function leaves that
function unchanged (see Sect. 9.3.7, above).
Ftr x Fx  2=x sin xT=2;

This shows that signal truncation incurs a 9.5.2 The Rectangular Window
smoothing effect due to convolution of the true (Box-Car)
spectrum F(x) with the spectrum of the window
function, W(x). In addition of the smoothing In order to get the least possible spectrum dis-
effect side lobes are also created. So, the spectrum tortion, the time interval of the window function
of a truncated signal is distorted, and the nature of must be as long and smooth (free of sharp cor-
distortion is the smoothing effect and develop- ners) as it can be (Bth 1974, pp. 155171).
ment of side lobes as shown by Fig. 9.20. Many types of window functions have been
Since an innitely long rectangular pulse has developed for application in this eld (8 window
a spectrum approaching an impulse function, it types were presented by Sheriff 2002, p. 397).

time-domain frequency-domain
truncation process convolution process
f(t) . w(t) F( ) * W( )

- R( ) * F( )

Fig. 9.20 Effect of varying the truncation window width (T) on smoothing of the spectrum F(x) of the truncated signal
f(t). Smoothing is more severe with shorter length, T
9.5 Spectra of Observational Data 217

A window designer usually aims at dening a 9.5.3 Triangular Window (Bartlett)


window function that has the least time-domain
distortion and at the same time incurs least This window, commonly known by the name
smoothing effect in the frequency domain. Two (Bartlett Window), is more often applied window
types of windows will be presented here: in seismic data processing. It is of a triangular
A window function w(t), is in the shape of a shape, dened as:
rectangular pulse of width (2T) and height (1). It
is dened as: wt 1  t=T jtj  T
0 jtj [ T
wt 1 jtj  T The spectrum window W(x), is square of the
0 jtj [ T spectrum of the rectangular function. It is given
by Papoulis (1962, p. 21): W(x) = (4/T) [sin
The spectrum of this window, given by (xT/2) / x]2 = T sinc2 (xT/2p).
Papoulis (1962, p. 20) is: W(x) = 2 (sin xT)/ Bartlett window has a tapering effect on the
x = 2T sinc(xT/p). truncated signal, introducing a change of a
This window is ideal in the time domain as it linear-scaling type. In the frequency domain,
isolates the function exactly as it is, with no any however, it causes relatively less spectrum dis-
modication. In the frequency domain, however, tortion (less smoothing) because of the sharp
it causes relatively high spectrum distortion impulsive shape of its spectrum. As it is indicated
(smoothing) because of the sharp corners which by its formula, Bartlett window (in the frequency
cause large side lobes. The spectrum window is a domain) has no negative values as the other
sinc-type function in which the width of its spectrum windows.
central lobe is inversely proportional to the width The two windows and their respective spectra
of the time-domain window. are sketched in Fig. 9.21.

(a) rectangular function, w (t) Fourier transform, W( )


2T
1

t - /T /T
-T 0 T - 0

(b) triangular function, w (t) Fourier transform, W( )

t
-T T - 2 /T 2 /T
Fig. 9.21 The two time windows in common use in seismic data processing and their spectra. a The triangular
time-function, or Box-car window and b the triangular time-function, or Bartlett window
218 9 The Seismic Processing Tools

9.6 Correlation Functions C12 ( ) = f1 ( ) f2 ( )


where, () represents correlation process, (s) is
9.6.1 Cross Correlation Function, called the time lag.
C12(s) In practice, it is a shift, multiply, and sum
process (Fig. 9.22).
Given two functions; f1(t) and f2(t), the
cross-correlation function C12(s) is dened as:
Z1 9.6.2 Autocorrelation Function
C12 s f1s  f2t sdt C11(s)
1
Given a function; f1(t), the autocorrelation
and, in digital summation over (i), it is: function C11(s) is dened as:
X
C12 j f 1 i  f 2 i j

for analogue functions: C11 () = f 1 () . f 1 ( t + ) dt

for digital functions: C11 (j) = f1 (i) . f 1 (i + j)

in symbolic form: C11 () = f 1 () f 1 ()

This denition can be represented in the fol- where, () represents correlation process, (s) is
lowing symbolical way: called the time lag.

f1 ( i ) = 2 2 1 -1

f2 ( i ) = 2 1 0 -1
2 1 0 -1
2 1 0 -1
2 1 0 -1
2 1 0 -1
2 1 0 -1
2 1 0 -1

C12 ( j ) = -2 -2 1 7 5 1 -2
f1 f2 C12

Fig. 9.22 Cross-correlation of the two functions (f1) and (f2)


9.6 Correlation Functions 219

Fig. 9.23 Autocorrelation


of the function (f1) f1 ( i ) = 2 2 1 -1

f1 ( i ) = 2 2 1 -1
2 2 1 -1
2 2 1 -1
2 2 1 -1
2 2 1 -1
2 2 1 -1
2 2 1 -1

C11 ( j ) = -2 0 5 10 5 0 -2

f1 f1 C11

In practice, it is a shift, multiply, and sum maxima at a period equal to the periodicity of the
process as it is done in computing the hidden periodic signal if it is existing in the orig-
cross-correlation function (Fig. 9.23). inal function. Most intensive use of the autocorre-
The function C11(s) is even function, lation functions is in computing power spectra and
C11(s) = C11(s), that is, symmetrical about the in the deconvolution computations.
point (s = 0).

9.6.4 Correlation Functions


9.6.3 Properties and Applications
in the Frequency
of Correlation
Domain
Functions
Correlation process, which is a shift-multiply-
The cross-correlation function serves as a tool to
and sum process, becomes a simple multiplica-
indicate the degree of similarity between the
tion process in the frequency domain, where the
correlated functions. Its peak grows large when
amplitude spectra of the involved functions are
the original functions are similar and decreases
multiplied and their respective phases subtracted,
otherwise. It approaches zero in case of random
as shown in the following table.
data as in case of random noise. Thus it can
detect signals embedded in random noise. Typi-
Time-domain Frequency-domain
cal application in seismic exploration is in sweep
f1(t) F1(x)
correlation with seismic traces obtained from
Vibroseis sources, and in computing the residual f2(t) F2(x)
static corrections. c12(t) = f1(t) f2(t) C12(x) = F1(x)  F2(x)
12(x) = 1(x) 2(x)
The autocorrelation function can be used in
detecting hidden periodicities. In this case c11(t) = f1(t) f1(t) C11(x) = {F1(x)}2
11(x) = 0
the autocorrelation function will show repeated
220 9 The Seismic Processing Tools

It is useful to note here that the autocorrelation function in the time domain is transformed into
power spectrum in the frequency domain. It is also important to note that the phase spectrum, 11(x),
is equal to zero. This provides an alternative method which allows computing power spectra from the
autocorrelation function instead of squaring the amplitude spectrum.

9.7 Convolution

Like correlation functions, Convolution is a mathematical process taking place between two functions
to output a new function the convolution output. Convolution of the two functions; f(t) and h(t) is
dened by the integral:

y(t) = f(t) * h(t) =

and, in digital summation over ( j), it is:

yj = fi * hi = fi . hj-i

The star symbol (*) represents convolution computation

Computation of the convolution process is done in the same way as in the correlation process
(described above) except for the need to reverse one of the two functions involved in the process.
Here is an example to clarify the process of convolving fi with hi where fi = { 2 1 3 0 1} and
hi = { 3 1 2}. The convolution process starts with reversing one of the two functions and then
continues with shifting, multiplication, and summing processes, shown as follows:

f( i ) = 2 1 3 0 -1

h(i)= 2 1 3
2 1 3
2 1 3
2 1 3
2 1 3
2 1 3
2 1 3

y (i) = 6 5 14 5 3 -1 -2

There are several other methods to compute convolution. These methods involve use of: special
tables, special matrix, and by z-transform. Methods of computation and mathematical properties of
the convolution process are summarized in Alsadi (1980, p. 8890).
The process involved in convolution (shift-multiply-and-sum process) is shown graphically in
Fig. 9.24, in which the function f(t) is convolved with the function h(t).
9.7 Convolution 221

f(t) (*) h(t) = y(t

t t
0 0

Fig. 9.24 Convolution of the functions f(t) and h(t)

In this example, h(t) is kept stationary while C11 t f 1 t  f 1 t


the function f(t) after being reversed, moved
along the time-axis, with the multiplication and These results gave another form of denition
summation processes. The output is a new of the cross-correlation and autocorrelation
function, y(t). functions in terms of convolution processes.
In the frequency domain, the convolution Convolution is the time-domain mathematical
between two time functions becomes multipli- computation that takes place in the process of
cation between their respective amplitude spectra frequency-ltering of signals.
and addition of the two phase spectra.

time -domain frequency -domain

f(t) F()
h(t) H()